point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

classic Classic list List threaded Threaded
22 messages Options
12
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Dr.MVH
After extensively searching the interwebs  I did not find a way to get the grasshopper3 USB3 camera to be recognized by micromanager. I tried to add the OpenCVgrabber in the hardware configuration, which worked with other usb2 cameras but that when the grasshopper is connected, adding the OpenCVgrabber crashes micromanager. The camera is recognized and works well with the flycapture 2 softwear from PGR. Is there a way to to circumvent this problem, either via converting frames to OpenCV and then using the OpenCVgrabber adapter in MM, or with another device adapter that will work generically with USB3 vision? For example Ximea offers a MM device adapter with their USB3 vision cameras so it should be possible. Looking forward to intelligent people helping out with this. Since USB3 is becoming increasingly used i am surprised that device adaptors for USB3 vision in MM are not more widespread. Profit-driven reasons perhaps for companies' own software and subscription-based support?

Thanks for any feedback.

Michel.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Dr.MVH
Dr.MVH wrote
 I did not find a way to get the grasshopper3 USB3 camera to be recognized by micromanager. I tried to add the OpenCVgrabber... in the hardware configuration, which worked with other usb2 cameras but when the grasshopper is connected, adding the OpenCVgrabber crashes micromanager. The camera is recognized and works well with the flycapture 2 softwear from PGR. Is there a way to to circumvent this problem, either via converting frames to OpenCV and then using the OpenCVgrabber adapter in MM, or with another device adapter that will work generically with USB3 vision?
This is an update and a question. My efforts took me to the JAI SDK (2.1.6) Win7 32-bit which now recognizes USB3-Vision devices. Indeed the camera (Grasshoper3 USB3) is recognized and works flawlessly in the JAI control tool. However the GigE device adapter for Micro-manager that makes use of the JAI SDK (1.4), will not load properly. On the other hand the JAI SDK (1.4) will not recognize USB3-Vision based cameras. Does anybody know whether the GigE device adapter was extended (or can be easily extended) to make use of the new JAI SDK (2.1.6) version? That would theoretically provide interfacing to Micro-manager for a lot of the cameras that use the USB3 Vision standard. Since I am not a programmer I do not know how easy it is to recompile the GigE device adapter to make use of the new JAI SDK.

Michel


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Sam Lord
Dr.MVH wrote
Does anybody know whether the GigE device adapter was extended (or can be easily extended) to make use of the new JAI SDK (2.1.6) version? That would theoretically provide interfacing to Micro-manager for a lot of the cameras that use the USB3 Vision standard. Since I am not a programmer I do not know how easy it is to recompile the GigE device adapter to make use of the new JAI SDK.
I don't know how to do that, but I do know that the JAI device in Micro-Manager only barely works with Point Grey GigE cameras, so I'm not surprised that it doesn't work with USB3 cameras.

Really, what we need to do is get Point Grey to recognize the value in writing device adapters for Micro-Manager. I'm mentioned it to them before, but I should do it again. And anyone else interested should do the same.

-Sam
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Kurt Thorn
On 3/7/2016 1:39 PM, Sam Lord wrote:

> Dr.MVH wrote
>> Does anybody know whether the GigE device adapter was extended (or can be
>> easily extended) to make use of the new JAI SDK (2.1.6) version? That
>> would theoretically provide interfacing to Micro-manager for a lot of the
>> cameras that use the USB3 Vision standard. Since I am not a programmer I
>> do not know how easy it is to recompile the GigE device adapter to make
>> use of the new JAI SDK.
> I don't know how to do that, but I do know that the JAI device in
> Micro-Manager only barely works with Point Grey GigE cameras, so I'm not
> surprised that it doesn't work with USB3 cameras.
>
> Really, what we need to do is get Point Grey to recognize the value in
> writing device adapters for Micro-Manager. I'm mentioned it to them before,
> but I should do it again. And anyone else interested should do the same.
>
> -Sam

Do you want to put together an open letter to Point Grey? I would be
happy to add my name.

Kurt

>
>
>
> --
> View this message in context: http://micro-manager.3463995.n2.nabble.com/point-grey-research-PGR-Grasshopper-3-USB3-vision-interfacing-with-MM-tp7586721p7586732.html
> Sent from the Micro-Manager mailing list archive at Nabble.com.
>
> ------------------------------------------------------------------------------
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://makebettercode.com/inteldaal-eval
> _______________________________________________
> micro-manager-general mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/micro-manager-general
>


--
Kurt Thorn
Associate Professor
Director, Nikon Imaging Center
http://thornlab.ucsf.edu/
http://nic.ucsf.edu/blog/


------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://makebettercode.com/inteldaal-eval
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

cherkas
In reply to this post by Sam Lord
I have tried some time ago writing device adapter for Grasshopper 3 USB3 but did not succeed due to my poor programming skills.
I am highly interested in making this work. I will push PointGrey again to realize importance of MicroManager for their products.

Best regards.

On Mon, Mar 7, 2016 at 11:39 PM, Sam Lord <[hidden email]> wrote:
Dr.MVH wrote
> Does anybody know whether the GigE device adapter was extended (or can be
> easily extended) to make use of the new JAI SDK (2.1.6) version? That
> would theoretically provide interfacing to Micro-manager for a lot of the
> cameras that use the USB3 Vision standard. Since I am not a programmer I
> do not know how easy it is to recompile the GigE device adapter to make
> use of the new JAI SDK.

I don't know how to do that, but I do know that the JAI device in
Micro-Manager only barely works with Point Grey GigE cameras, so I'm not
surprised that it doesn't work with USB3 cameras.

Really, what we need to do is get Point Grey to recognize the value in
writing device adapters for Micro-Manager. I'm mentioned it to them before,
but I should do it again. And anyone else interested should do the same.

-Sam



--
View this message in context: http://micro-manager.3463995.n2.nabble.com/point-grey-research-PGR-Grasshopper-3-USB3-vision-interfacing-with-MM-tp7586721p7586732.html
Sent from the Micro-Manager mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://makebettercode.com/inteldaal-eval
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general


------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://makebettercode.com/inteldaal-eval
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Dr.MVH
In reply to this post by Kurt Thorn

> Dr.MVH wrote
>> Does anybody know whether the GigE device adapter was extended (or can be
>> easily extended) to make use of the new JAI SDK (2.1.6) version? That
>> would theoretically provide interfacing to Micro-manager for a lot of the
>> cameras that use the USB3 Vision standard. Since I am not a programmer I
>> do not know how easy it is to recompile the GigE device adapter to make
>> use of the new JAI SDK.

>> Michel


> Really, what we need to do is get Point Grey to recognize the value in
> writing device adapters for Micro-Manager. I'm mentioned it to them before,
> but I should do it again. And anyone else interested should do the same.
>
> -Sam

Do you want to put together an open letter to Point Grey? I would be
happy to add my name.

Kurt

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general



Dear Sam,
Dear Kurt,
I have already contacted Point Grey and their answer was that "gathering from their customer feedback their cameras are not compatible with this tool (MicroManager) and therefore support can not be extended..." I would be very interested in convincing them to compile a MicroManager device adapter for their cameras. However, I also think that a general MicroManager device adapter for USB3-Vision cameras should exist.  I am currently in contact with Point Grey, looking into some alternative ways. I will be keeping this thread up to date with the developments as to help other interested individuals in  PGR-MicroManager interfacing. Let´s wait till the end of the week for more parties willing to join an open letter effort to Point Grey for a MicroManager device adapter.

As a side note, I have my reservations about the new business model of the MicroManager developing team at Open Imaging and the subscription-based model. Perhaps companies like Point Grey and others that profit mainly from camera and equipment sales should support Open Imaging in developing interfacing with their products. Unfortunately, I noticed that these companies are often interested in business partners that offer closed source software for their products, apparently for increasing profits. Your feedback is very welcome.

Michel
 





Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Nico Stuurman-2

On 3/8/16 1:16 PM, Dr.MVH wrote:

> Ihave already contacted Point Grey and their answer was that
> "gathering from their customer feedback their cameras are not
> compatible with this tool (MicroManager) and therefore support can not
> be extended..." I would be very interested in convincing them to
> compile a MicroManager device adapter for their cameras. However, I
> also think that a general MicroManager device adapter for USB3-Vision
> cameras should exist. I am currently in contact with Point Grey,
> looking into some alternative ways. I will be keeping this thread up
> to date with the developments as to help other interested individuals
> in PGR-MicroManager interfacing. Let´s wait till the end of the week
> for more parties willing to join an open letter effort to Point Grey
> for a MicroManager device adapter. As a side note, I have my
> reservations about the new business model of the MicroManager
> developing team at Open Imaging and the subscription-based model.

Curious to hear what your reservations are and what other models you
propose to finance development and maintenance of freely available, open
source software for microscope control.

> Perhaps companies like Point Grey and others that profit mainly from
> camera and equipment sales should support Open Imaging in developing
> interfacing with their products.
I certainly hope they would (and some of them do with more than just
developing interfaces to their hardware, if that was all support there
would still be a big gap to cover consisting of integration and GUI
development amongst others), and I applaud your efforts to convince
Point Grey to develop a device adapter for Micro-Manager, but I also
understand their point of view (i.e. the sales numbers may not warrant
the development effort).  But in the mean time, why not look at similar
cameras from companies that have written a Micro-Manager device adapter
and are supporting their code, such as ABS
(http://www.kameras.abs-jena.de/home_en.html) and Ximea
(https://www.ximea.com)?

> Unfortunately, I noticed that these companies are often interested in
> business partners that offer closed source software for their
> products, apparently for increasing profits.
I think that increasing profits is the mission of most companies (other
than Open Imaging;)

Best,

Nico


------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Dr.MVH
Dear Nico,
First of all let me say that I am sure the scientific community (including myself) is grateful to you and your colleagues for the development and maintenance of MicroManager. My answers to your comments below are in no way aimed to undermine this amazing tool and your work.  

Nico Stuurman-2 wrote
>Curious to hear what your reservations are and what other models you
>propose to finance development and maintenance of freely available, open
>source software for microscope control.

My reservations have to do with consumer psychology. A tool that used to be free (and still is), suddenly turns into a subscription-based business model. According to Open Imaging "Open Imaging is committed to keeping µManager free and open source. However, the development and distribution of the software will be supported by paid services that we provide". For many consumers this could also translate as : come on guys, we´ve been providing free open forum-support for years but now we will charge for it (because NIH can not provide two postdoc positions). However the forum will continue existing (but we will not be providing "valuable answers" to questions that can instead make money under a subscription umbrella. Say some laboratories understand the financial gains and take the subscription route (including myself and my suggestions to the departments i am associated with), while others do not and they either try to exploit the open-sourceness of MM themselves or abandon a free tool and instead go for a paid, apparently  "legit" tool that comes with all the support they could get. What would be the result? A few labs that get the juicy insider information via subscription and a lot of labs that abandon MM altogether, thus a reduction in MM user base. In a nutshell, my opinion is that people will find it hard to pay for something that used to be free (and that was a major appeal of MM). Content creators on Youtube are not receiving money directly from viewers (consumers) by paid subscription. Instead, they receive money from advertising based on the size of their viewer base. I think that the MM team should figure out a way to get the funds from the organizations that make lots of it and focus on increasing their user base.  

 

>I certainly hope they would (and some of them do with more than just
>developing interfaces to their hardware, if that was all support there
>would still be a big gap to cover consisting of integration and GUI
>development amongst others), ...

I was referring to direct financial support to Open Imaging by other companies. i.e paying you to develop and integrate their devices with micromanager. They could then use the resultant interfacing as a selling incentive to end customers.

>and I applaud your efforts to convince
>Point Grey to develop a device adapter for Micro-Manager, but I also
>understand their point of view (i.e. the sales numbers may not warrant
>the development effort).  

Perhaps if they would advertise the compatibility with Micro-Manager their sales numbers would increase (someone needs to offer the incentive first and customers will follow). In addition, although I have no idea about programing, I can envision that given the standardization of USB3vision technology, an experienced programer could "reverse engineer" the Ximea device adapter and recompile it using the FlyCapture2 SDK from Point Grey relatively easily (or perhaps I am very naive).


>But in the mean time, why not look at similar
>cameras from companies that have written a Micro-Manager device adapter
>and are supporting their code, such as ABS
>(http://www.kameras.abs-jena.de/home_en.html) and Ximea
>(https://www.ximea.com)?

ABS does not provide details about the sensors used in their cameras and Ximea does not offer a camera with the ICX825 sensor that I know would fit the cell biology needs of our particular set up. In addition, so far Point Grey offers the best value for money (see below).


>I think that increasing profits is the mission of most companies (other
>than Open Imaging;)

Transiently increasing profits should take second place to reaching a sustained monopoly. As an example, I recently asked an systems integrator for a quote. I described that we have an old filter and shutter controller and a MAC5000 stage and focus controller fitted on an old Axiovert 200 (non motorized) microscope. We would need the integration plus a USB3 camera based on the ICX825 sensor. Their quote was close to 15,000 €!!!!. Instead we used Micro-Manager and within a week these old components (microscope, filter, shutter and stage) work better than i ever thought. We then purchased the camera in question (Grasshopper 3 USB3) for 1200€. Do you think that this unnamed systems integrator has a good standing in my eyes or the departments i am associated with and a diverse cell biology circle around the University? I´m more willing to go around the University converting people to MM than fall victim to exploitation. If Open Imaging finds a way to highlight the merits and financial gains of MM then you could create a justifiable monopoly (and I totally applaud that).

Best,
Michel


------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

JonD
Administrator
Hi Michel and others, permit me to comment on a couple of your points.

Dr.MVH wrote
My reservations have to do with consumer psychology. A tool that used to be free (and still is), suddenly turns into a subscription-based business model. According to Open Imaging "Open Imaging is committed to keeping µManager free and open source. However, the development and distribution of the software will be supported by paid services that we provide". For many consumers this could also translate as : come on guys, we´ve been providing free open forum-support for years but now we will charge for it (because NIH can not provide two postdoc positions). However the forum will continue existing (but we will not be providing "valuable answers" to questions that can instead make money under a subscription umbrella. Say some laboratories understand the financial gains and take the subscription route (including myself and my suggestions to the departments i am associated with), while others do not and they either try to exploit the open-sourceness of MM themselves or abandon a free tool and instead go for a paid, apparently  "legit" tool that comes with all the support they could get. What would be the result? A few labs that get the juicy insider information via subscription and a lot of labs that abandon MM altogether, thus a reduction in MM user base. In a nutshell, my opinion is that people will find it hard to pay for something that used to be free (and that was a major appeal of MM). Content creators on Youtube are not receiving money directly from viewers (consumers) by paid subscription. Instead, they receive money from advertising based on the size of their viewer base. I think that the MM team should figure out a way to get the funds from the organizations that make lots of it and focus on increasing their user base.
The software is still free and open-source.  Users can still help each other via this forum.  But if you want the "true experts" to help you personally or if you want a specific feature developed then that is no longer free because somebody has to pay for it and NIH isn't paying for it anymore.

Continuing serious development requires least one or two people devoting their full time efforts to improving the core program besides device adapters.  NIH made it clear that they would not financially support Micro-Manager past 2015.  So what to do?  Because some labs are willing to pay for support, Open Imaging sells that in order to fund ongoing development.  Importantly the fruits of ongoing development are shared by everybody whether or not they are paying Open Imaging.  To me this is best solution possible given the circumstances.  And it's not unlike the way Linux development is funded.


Dr.MVH wrote
I was referring to direct financial support to Open Imaging by other companies. i.e paying you to develop and integrate their devices with micromanager. They could then use the resultant interfacing as a selling incentive to end customers.
I'm pretty sure Open Imaging accepts money from companies to develop Micro-Manager device adapters.  Many companies (like mine) would prefer to develop device adapters themselves, in this way they achieve the same result via different means (i.e. they pay their employees to do the work instead of paying Open Imaging).

Having (working) device adapters can definitely can be a selling point for hardware no matter how they get created.  But for some companies the marginal sales may not be worth the expense.  In which case the end user (e.g. you) are free to pay Open Imaging to develop a device adapter or else spend the time to create it yourself.  Of course many will just "vote with their feet" and buy hardware from companies that are providing and supporting device adapters for their hardware either directly or via Open Imaging's services.

Jon

------------------------------------------
Jon Daniels
Applied Scientific Instrumentation
29391 West Enid Rd, Eugene, OR 97402
Phone: (541) 461-8181 x118
------------------------------------------
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Dr.MVH
In reply to this post by Dr.MVH
Dr.MVH wrote
Dear Nico,
First of all let me say that I am sure the scientific community (including myself) is grateful to you and your colleagues for the development and maintenance of MicroManager. My answers to your comments below are in no way aimed to undermine this amazing tool and your work.  

Nico Stuurman-2 wrote
>Curious to hear what your reservations are and what other models you
>propose to finance development and maintenance of freely available, open
>source software for microscope control.

My reservations have to do with consumer psychology. A tool that used to be free (and still is), suddenly turns into a subscription-based business model. According to Open Imaging "Open Imaging is committed to keeping µManager free and open source. However, the development and distribution of the software will be supported by paid services that we provide". For many consumers this could also translate as : come on guys, we´ve been providing free open forum-support for years but now we will charge for it (because NIH can not provide two postdoc positions). However the forum will continue existing (but we will not be providing "valuable answers" to questions that can instead make money under a subscription umbrella. Say some laboratories understand the financial gains and take the subscription route (including myself and my suggestions to the departments i am associated with), while others do not and they either try to exploit the open-sourceness of MM themselves or abandon a free tool and instead go for a paid, apparently  "legit" tool that comes with all the support they could get. What would be the result? A few labs that get the juicy insider information via subscription and a lot of labs that abandon MM altogether, thus a reduction in MM user base. In a nutshell, my opinion is that people will find it hard to pay for something that used to be free (and that was a major appeal of MM). Content creators on Youtube are not receiving money directly from viewers (consumers) by paid subscription. Instead, they receive money from advertising based on the size of their viewer base. I think that the MM team should figure out a way to get the funds from the organizations that make lots of it and focus on increasing their user base.  

 

>I certainly hope they would (and some of them do with more than just
>developing interfaces to their hardware, if that was all support there
>would still be a big gap to cover consisting of integration and GUI
>development amongst others), ...

I was referring to direct financial support to Open Imaging by other companies. i.e paying you to develop and integrate their devices with micromanager. They could then use the resultant interfacing as a selling incentive to end customers.

>and I applaud your efforts to convince
>Point Grey to develop a device adapter for Micro-Manager, but I also
>understand their point of view (i.e. the sales numbers may not warrant
>the development effort).  

Perhaps if they would advertise the compatibility with Micro-Manager their sales numbers would increase (someone needs to offer the incentive first and customers will follow). In addition, although I have no idea about programing, I can envision that given the standardization of USB3vision technology, an experienced programer could "reverse engineer" the Ximea device adapter and recompile it using the FlyCapture2 SDK from Point Grey relatively easily (or perhaps I am very naive).


>But in the mean time, why not look at similar
>cameras from companies that have written a Micro-Manager device adapter
>and are supporting their code, such as ABS
>(http://www.kameras.abs-jena.de/home_en.html) and Ximea
>(https://www.ximea.com)?

ABS does not provide details about the sensors used in their cameras and Ximea does not offer a camera with the ICX825 sensor that I know would fit the cell biology needs of our particular set up. In addition, so far Point Grey offers the best value for money (see below).


>I think that increasing profits is the mission of most companies (other
>than Open Imaging;)

Transiently increasing profits should take second place to reaching a sustained monopoly. As an example, I recently asked an systems integrator for a quote. I described that we have an old filter and shutter controller and a MAC5000 stage and focus controller fitted on an old Axiovert 200 (non motorized) microscope. We would need the integration plus a USB3 camera based on the ICX825 sensor. Their quote was close to 15,000 €!!!!. Instead we used Micro-Manager and within a week these old components (microscope, filter, shutter and stage) work better than i ever thought. We then purchased the camera in question (Grasshopper 3 USB3) for 1200€. Do you think that this unnamed systems integrator has a good standing in my eyes or the departments i am associated with and a diverse cell biology circle around the University? I´m more willing to go around the University converting people to MM than fall victim to exploitation. If Open Imaging finds a way to highlight the merits and financial gains of MM then you could create a justifiable monopoly (and I totally applaud that).

Best,
Michel


------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Dr.MVH
In reply to this post by JonD
Hi Jon and I apologize to you and the forum for the previous empty reply. My bad and my keyboard's. Although the purpose of my original post was just seeking advice on interfacing USB3Vision Point Grey camera with Micro-Manager, the discussion took a more business-related direction (not my forte since I´m a cell biologist). However, you will find my (no expert) comments below:

JonD wrote
Hi Michel and others, permit me to comment on a couple of your points.

Importantly the fruits of ongoing development are shared by everybody whether or not they are paying Open Imaging.  To me this is best solution possible given the circumstances.  And it's not unlike the way Linux development is funded...
What is the incentive for those few that pay so the majority can enjoy the fruits?

JonD wrote
I'm pretty sure Open Imaging accepts money from companies to develop Micro-Manager device adapters.  Many companies (like mine) would prefer to develop device adapters themselves, in this way they achieve the same result via different means (i.e. they pay their employees to do the work instead of paying Open Imaging).
Wouldn´t that qualify as outsourcing Open Imaging at the expense of the end user? On the premise that an in-house developed device adapter is of market value to many companies (including yours) as you say, then the existence of Open Imaging and MicroManager is automatically of value to you. Shouldn´t that be a reason to justify a little cut from your profits to Open Imaging, depending whether the end user shows interest in using your technology with MicroManager?


JonD wrote
Having (working) device adapters can definitely can be a selling point for hardware no matter how they get created.  But for some companies the marginal sales may not be worth the expense.
Is there really a high expense involved in writing a device adapter for a standardized USB3 technology, using standardized imaging sensors, in standardized programming language when an experienced programmer has access to the SDK provided by the vendor?


JonD wrote
  In which case the end user (e.g. you) are free to pay Open Imaging to develop a device adapter or else spend the time to create it yourself.  Of course many will just "vote with their feet" and buy hardware from companies that are providing and supporting device adapters for their hardware either directly or via Open Imaging's services.

Jon
If the end user had an option to purchase either a vendor's camera that supports Micro-Manager for 4500 or an identical camera (i.e identical sensor, albeit with different housing and stickers) from another vendor without Micro-Manager support for 1000, which option would carry more probability, assuming the end user is your average biologist that does a bit of basic googling?  Perhaps a lot (vast majority) of users will prefer the 1000 camera and would go via DirectShow and OpenCVgrabber, even if the controls are not making full use of the cameras capabilities but would suffice for certain experimental set ups.
Thanks for your comments.
Regards,
Michel
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Sam Lord
Dr.MVH wrote
What is the incentive for those few that pay so the majority can enjoy the fruits?
It's very simple: think of it as a free version and a premium version. You get the stable program for free. If you want nightly builds or you want direct support from the Open Imaging team, you pay for the premium version. Many programs work like this, such as Evernote. It's still unclear whether this business model will keep Open Imaging afloat, but it's a reasonable approach that succeeds in keeping Micro-Manager open and free and community-driven.

Dr.MVH wrote
JonD wrote
I'm pretty sure Open Imaging accepts money from companies to develop Micro-Manager device adapters.  Many companies (like mine) would prefer to develop device adapters themselves, in this way they achieve the same result via different means (i.e. they pay their employees to do the work instead of paying Open Imaging).
Wouldn´t that qualify as outsourcing Open Imaging at the expense of the end user?
I'm not sure what your concern is. A company can either write its own device adapter or pay some other programers to do it (e.g. Open Imaging). Either way, users get a new device adapter for free. Or you can write it yourself or hire some programmer to. No one is stopping you!

Dr.MVH wrote
Is there really a high expense involved in writing a device adapter for a standardized USB3 technology, using standardized imaging sensors, in standardized programming language when an experienced programmer has access to the SDK provided by the vendor?
My experience with the GigE adapter reveals that one adapter does not necessarily work for all camera brands, even if they do use the same "standard." Each camera may have different settings that the adapter would need to control in different ways. Anyway, what "experienced programmer" are you proposing performs this service for free? Why don't you hire someone to do this? There's no free lunch. Someone has to spend their time working on this problem.

Dr.MVH wrote
If the end user had an option to purchase either a vendor's camera that supports Micro-Manager for 4500 or an identical camera (i.e identical sensor, albeit with different housing and stickers) from another vendor without Micro-Manager support for 1000, which option would carry more probability, assuming the end user is your average biologist that does a bit of basic googling?  Perhaps a lot (vast majority) of users will prefer the 1000 camera and would go via DirectShow and OpenCVgrabber, even if the controls are not making full use of the cameras capabilities but would suffice for certain experimental set ups.
Well, you get what you pay for: If you don't want to pay more for a camera that supports Micro-Manager, you won't get Micro-Manager support.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Nico Stuurman-2
In reply to this post by Dr.MVH

On 3/10/16 11:21 AM, Dr.MVH wrote:
> First of all let me say that I am sure the scientific community (including
> myself) is grateful to you and your colleagues for the development and
> maintenance of MicroManager. My answers to your comments below are in no way
> aimed to undermine this amazing tool and your work.
I also want to start by thanking you for sharing your thoughts, it is so
much better to have those out in the open rather than percolating
through the community.  Mark and Chris (with my and Ron's help) have
setup Open Imaging as a way to continue to provide open source
microscope imaging tools for the scientific community. It is actually
quite difficult to fund a project like this (believe me, I have been
writing grants for this for over a decade), and I think that Open
Imaging came up with am exciting model, providing high quality support
to scientists who choose to pay for support, as well as support by
companies (a model that is slowly developing but that you will surely
hear more about in the future), at the same time making the software
freely available.  If successful (and the first half a year was
successful and the outlook is good), then this would also become a great
model for development of scientific open source software, something that
many of us desire and/or demand (certainly if one reads the editorials
of Nature Methods).

I hope that sets the spirit of Open Imaging straight, there are a few
points that came up that I would like to address specifically:

Dr.MVH wrote:
>> My reservations have to do with consumer psychology. A tool that used
>> to be free (and still is), suddenly turns into a subscription-based
>> business model. According to Open Imaging "Open Imaging is committed
>> to keeping µManager free and open source. However, the development
>> and distribution of the software will be supported by paid services
>> that we provide". For many consumers this could also translate as :
>> come on guys, we´ve been providing free open forum-support for years
>> but now we will charge for it (because NIH can not provide two
>> postdoc positions).

I think you are playing devil's advocate here, since this section is
full of mistrust, however, one fact: Micro-Manager was never executed by
two post-docs, rather by professional programmers (who cost a lot more,
certainly in San Francisco).  Funding levels by NIH are publicly
visible, and can be seen here:
https://projectreporter.nih.gov/project_info_history.cfm?aid=7619894&icde=8337770 
(and I am pretty sure that Open Imaging operates a lot leaner, in part
since there is so much less overhead).

>> However the forum will continue existing (but we will not be
>> providing "valuable answers" to questions that can instead make money
>> under a subscription umbrella.

I have not noticed such a trend, and again, to even think that way is to
start out by distrusting the people doing the work (and if you do not
trust them, why bother having anything to do with it at all?)

>> Say some laboratories understand the financial gains and take the
>> subscription route (including myself and my suggestions to the
>> departments i am associated with), while others do not and they
>> either try to exploit the open-sourceness of MM themselves or abandon
>> a free tool and instead go for a paid, apparently "legit" tool that
>> comes with all the support they could get.

I do not understand this choice.  If a lab wants to do the support
themselves (and is not interested in assuring future development), then
go ahead and use MM.  If a lab wants a fully supported tool, then MM is
by far the most economical solution (very "legit", and on top of it open
source with all its advantages).

>> What would be the result? A few labs that get the juicy insider
>> information via subscription and a lot of labs that abandon MM
>> altogether, thus a reduction in MM user base. In a nutshell, my
>> opinion is that people will find it hard to pay for something that
>> used to be free (and that was a major appeal of MM).

Luckily, many labs disagree with this opinion, and are interested in
high level support and willing to pay for it.  On top of that, Open
Imaging keeps MM freely available!  So, I really, really do not
understand this sentiment (and do not see any evidence of labs
abandoning MM).

>> Content creators on Youtube are not receiving money directly from
>> viewers (consumers) by paid subscription. Instead, they receive money
>> from advertising based on the size of their viewer base. I think that
>> the MM team should figure out a way to get the funds from the
>> organizations that make lots of it and focus on increasing their user
>> base.

The difference in market size between YouTube/phone apps and scientific
microscope software is several orders of magnitude, and funding models
simply do not translate.  I totally agree that it would be great if
companies making hardware for microscopy would help to sustain MM.  Some
already do, and there will be progress on that front, but even those
companies will want to see that scientists really value MM, and the best
way of showing that is by purchasing a support contract.

>> Perhaps if Point Grey would advertise the compatibility with
>> Micro-Manager their sales numbers would increase (someone needs to
>> offer the incentive first and customers will follow).
Initial development of a camera adapter for MM costs ~$10-20k (estimate
depending on complexity).  If you think that is a wise investment, it
should be easy to convince Point Grey.
>> In addition, although I have no idea about programing, I can envision
>> that given the standardization of USB3vision technology, an
>> experienced programer could "reverse engineer" the Ximea device
>> adapter and recompile it using the FlyCapture2 SDK from Point Grey
>> relatively easily (or perhaps I am very naive).
As a (part-time) programmer, I do not understand what you mean.  I can
tell you however, that writing a USB3vision adapter for MM will likely
be an order of magnitude harder then writing directly against a
company's SDK (so, think $100k-$200k).  I wrote the initial dc1394
("firewire") adapter exactly with this idea in mind: one adapter that
can work with all cameras that adhere to the specs.  It turned out that
the specs allow for so many variations, that it is virtually impossible
to anticipate all possible permutations, on top of that, many vendor
have variations around the specs.  Mark did a much better job with a
second version of the dc1394 adapter, but it still will only work with a
subset of cameras.  The situation with dc1394 was much better than with
USB3Vision, since there actually is an open source, cross-platform
library (libdc1394: https://sourceforge.net/projects/libdc1394/).  There
is no such thing for USB3Vision (for instance:
http://stackoverflow.com/questions/29653529/will-there-ever-be-a-libdc1394-like-api-for-usb3-vision-and-or-gige-vision-camer),
so writing a MM USB3Vision adapter would involve first writing the low
level code on all supported platforms (Mac, Windows, Linux), further
driving up the cost for development.

Another general issue is that the software driving machine vision
cameras was all written without awareness of requirements for
microscopy.  Microscopists value accurately timed beginning and end of a
"snap", so that shutters can be synchronized with exposure. Most machine
vision camera's software does not let you accurately time those things.  
Writing control software that does this right takes time and requires
expertise, hence rises the cost.  I expect that in the near future, most
of the cost for microscope cameras is in software rather than hardware
(this may already be the case even when we end-users do not perceive it
that way).

Hope this enlightens a bit!

Best,

Nico

------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Mark Tsuchida-3
Hi Michel, and µManager Community,

As CEO of Open Imaging, I would like to thank you for expressing your
opinions about the µManager project. As some of you have noted, our
most important goal is to continue the success of µManager as a
community-oriented project, and we always value your input.

On Thu, Mar 10, 2016 at 11:21 AM, Dr.MVH <[hidden email]> wrote:

> Nico Stuurman-2 wrote
> >Curious to hear what your reservations are and what other models you
> >propose to finance development and maintenance of freely available, open
> >source software for microscope control.
>
> My reservations have to do with consumer psychology. A tool that used to
> be free (and still is), suddenly turns into a subscription-based business
> model. According to Open Imaging "Open Imaging is committed to keeping
> µManager free and open source. However, the development and distribution
> of the software will be supported by paid services that we provide". For
> many consumers this could also translate as : come on guys, we've been
> providing free open forum-support for years but now we will charge for it
> (because NIH can not provide two postdoc positions). However the forum
> will continue existing (but we will not be providing "valuable answers" to
> questions that can instead make money under a subscription umbrella. Say
> some laboratories understand the financial gains and take the subscription
> route (including myself and my suggestions to the departments i am
> associated with), while others do not and they either try to exploit the
> open-sourceness of MM themselves or abandon a free tool and instead go for
> a paid, apparently  "legit" tool that comes with all the support they
> could get. What would be the result? A few labs that get the juicy insider
> information via subscription and a lot of labs that abandon MM altogether,
> thus a reduction in MM user base. In a nutshell, my opinion is that people
> will find it hard to pay for something that used to be free (and that was
> a major appeal of MM). Content creators on Youtube are not receiving money
> directly from viewers (consumers) by paid subscription. Instead, they
> receive money from advertising based on the size of their viewer base. I
> think that the MM team should figure out a way to get the funds from the
> organizations that make lots of it and focus on increasing their user
> base.

There is no denying that it is hard to ask for money for something
that has been free -- although some µManager users (and developers)
have very kindly signed up for our service program early on; without
this we would have been bankrupt already! (So thanks goes to those
labs and companies.)

But I think Sam's characterization (freemeum model) is a good way to
think about this. It should be noted that the µManager user base is
much larger (~10-fold?) than the subscribers of this mailing list.
Presumably these users are finding µManager useful enough without
getting support (at least not from us). And in the vast majority of
cases, our service program pricing is very competitive with
proprietary software, so I don't think this by itself should
discourage use of µManager if we do our job and keep up with the
development.

I should also mention that, as the user base has gotten larger and as
people have started doing diverse things with µManager, answering
every single question that comes our way (e.g. via this mailing list)
had already started to become unsustainable before we transitioned to
Open Imaging. We would need a team of several engineers to handle all
of those. We hope that the paid support model will allow us to scale
our team with demand as more people subscribe.


> >I certainly hope they would (and some of them do with more than just
> >developing interfaces to their hardware, if that was all support there
> >would still be a big gap to cover consisting of integration and GUI
> >development amongst others), ...
>
> I was referring to direct financial support to Open Imaging by other
> companies. i.e paying you to develop and integrate their devices with
> micromanager. They could then use the resultant interfacing as a selling
> incentive to end customers.

This is something we might do as part of our business, but, as others
have pointed out, µManager development is not just about adding device
support. For this model to work on its own, we would have to charge a
very high price for development of device support in order to
subsidize development of the core program and user support.


On Thu, Mar 10, 2016 at 4:48 PM, Dr.MVH <[hidden email]> wrote:

> JonD wrote
>> I'm pretty sure Open Imaging accepts money from companies to develop
>> Micro-Manager device adapters.  Many companies (like mine) would prefer to
>> develop device adapters themselves, in this way they achieve the same
>> result via different means (i.e. they pay their employees to do the work
>> instead of paying Open Imaging).
>
> Wouldn't that qualify as outsourcing Open Imaging at the expense of the end
> user? On the premise that an in-house developed device adapter is of market
> value to many companies (including yours) as you say, then the existence of
> Open Imaging and MicroManager is automatically of value to you. Shouldn't
> that be a reason to justify a little cut from your profits to Open Imaging,
> depending whether the end user shows interest in using your technology with
> MicroManager?

We hope so, too, and for that reason we are also starting to offer
services (even before development per se) and sponsorship
opportunities to device manufacturers. This will become a little more
visible once we have an official µManager release from Open Imaging
(we're working very hard on this right now).

On the other hand, we don't want to be "owned" solely by (a subset of)
device manufacturers, either. I think the µManager community derives
its strength from participation from both scientists/users and device
manufacturers, and it makes sense for us to offer paid benefits to
both of those groups.

(Also, the "shouldn't that be a reason to justify a little cut"
argument can be applied to well-funded scientists, too, if they are
saving money by not buying proprietary software :)


> JonD wrote
>> Having (working) device adapters can definitely can be a selling point for
>> hardware no matter how they get created.  But for some companies the
>> marginal sales may not be worth the expense.
>
> Is there really a high expense involved in writing a device adapter for a
> standardized USB3 technology, using standardized imaging sensors, in
> standardized programming language when an experienced programmer has access
> to the SDK provided by the vendor?

Very roughly speaking, probably a few to several tens of thousands of
dollars, if you'd like to know (camera adapters are the most complex).
Of course it depends on many details. "Standard" interfaces are more
work (to get right) than vendor-specific ones, because they need to
work with a wider range of devices.

Actually I've always hoped for µManager to have standard GenICam
(USB3Vision/GigEVision) support, it's just that it's a relatively
major project and never has been high enough on the list of
high-priority tasks.

I can imagine that for a company selling relatively inexpensive
cameras (therefore in large quantities, and primarily outside of
microscopy), additional sales created by µManager support would not be
attractive enough.


> JonD wrote
>>   In which case the end user (e.g. you) are free to pay Open Imaging to
>> develop a device adapter or else spend the time to create it yourself.  Of
>> course many will just "vote with their feet" and buy hardware from
>> companies that are providing and supporting device adapters for their
>> hardware either directly or via Open Imaging's services.
>>
>> Jon
>
> If the end user had an option to purchase either a vendor's camera that
> supports Micro-Manager for 4500 or an identical camera (i.e identical
> sensor, albeit with different housing and stickers) from another vendor
> without Micro-Manager support for 1000, which option would carry more
> probability, assuming the end user is your average biologist that does a bit
> of basic googling?  Perhaps a lot (vast majority) of users will prefer the
> 1000 camera and would go via DirectShow and OpenCVgrabber, even if the
> controls are not making full use of the cameras capabilities but would
> suffice for certain experimental set ups.

Maybe. It depends on what 'typical' is (a dangerous word to apply to
scientists). µManager was originally developed in a lab that does
serious microscopy, and initially adopted in similar labs (think:
cameras costing $10k-50k). It's great that we have gradually also come
to support less expensive cameras, vastly expanding our potential user
base, but (as I have previously noted on this mailing list) the
question of who should pay for development of support for the least
expensive range of cameras remains a hard one, because in most cases
nobody makes a huge amount of money from it.

In fact, we (Open Imaging) are trying to be the entity that works on
the important aspects of µManager that otherwise nobody would work on.
Support for standard camera interfaces is the sort of thing that we
might want to tackle ourselves maybe if we get enough subscriptions in
the future to expand our team.

In the short term, with our limited resources, we will likely need to
concentrate on higher-priority tasks in the core program, unless
somebody wants to specifically pay us to work on it, unfortunately.

Best regards,
Mark


Mark A. Tsuchida
CEO & Co-CTO
Open Imaging, Inc.
https://open-imaging.com

------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Austin
I think it’s important to note that mass marketing techniques employed by consumer products (such as youtube videos) are tough to apply for our community.
1. They have an exponentially larger possible customer base.
2. They are backed by large multinational organizations.

This is a small market.
Small market = less customers.
But the guys maintaining and supporting the code need to eat.
Less customers = higher cost per customer for the supporting cast to eat.

It’s either a subscription service, or the MM guys start either selling hardware and taking a %, or straight charging per hour for labor.

Of the three possibilities above, I think the subscription service is the most open/fair.

I’d also note that the alternative is…..extremely expensive closed-source and high subscription cost commercial products.

If anyone has a better solution we should chime in, as now is the time! But I think this is a case of “best balanced solution” that’s been proven to work in other arenas.

The only other case I can think of where a truly open project worked was via corporate funding - something like Reptier. Even that isn’t as applicable as the install base for 3Dprinters is some orders of magnitude > scope installs.

Consider some thousand multiplier of install bases globally, what, 5k, * $100/user/year, = $500k.
$500k for facilities, transport, marketing, taxation, healthcare, disability, insurance, comms, etc etc. come to what? $100k per programmer pay? When these gents could easily work for a firm in the bay area that pays 50% more, with better hours?  

I think we need to be honest about what our community wants to see. If we choke off the funding for MM it’ll disappear, and we’ll be left with higher cost, closed source options.

For comparison, a “multidimensional capture” renewal license for MetaMorph or Elements runs around $750-$1,400. EVERY YEAR….

Austin

> On Mar 11, 2016, at 12:18 PM, Mark Tsuchida <[hidden email]> wrote:
>
> Hi Michel, and µManager Community,
>
> As CEO of Open Imaging, I would like to thank you for expressing your
> opinions about the µManager project. As some of you have noted, our
> most important goal is to continue the success of µManager as a
> community-oriented project, and we always value your input.
>
> On Thu, Mar 10, 2016 at 11:21 AM, Dr.MVH <[hidden email]> wrote:
>> Nico Stuurman-2 wrote
>>> Curious to hear what your reservations are and what other models you
>>> propose to finance development and maintenance of freely available, open
>>> source software for microscope control.
>>
>> My reservations have to do with consumer psychology. A tool that used to
>> be free (and still is), suddenly turns into a subscription-based business
>> model. According to Open Imaging "Open Imaging is committed to keeping
>> µManager free and open source. However, the development and distribution
>> of the software will be supported by paid services that we provide". For
>> many consumers this could also translate as : come on guys, we've been
>> providing free open forum-support for years but now we will charge for it
>> (because NIH can not provide two postdoc positions). However the forum
>> will continue existing (but we will not be providing "valuable answers" to
>> questions that can instead make money under a subscription umbrella. Say
>> some laboratories understand the financial gains and take the subscription
>> route (including myself and my suggestions to the departments i am
>> associated with), while others do not and they either try to exploit the
>> open-sourceness of MM themselves or abandon a free tool and instead go for
>> a paid, apparently  "legit" tool that comes with all the support they
>> could get. What would be the result? A few labs that get the juicy insider
>> information via subscription and a lot of labs that abandon MM altogether,
>> thus a reduction in MM user base. In a nutshell, my opinion is that people
>> will find it hard to pay for something that used to be free (and that was
>> a major appeal of MM). Content creators on Youtube are not receiving money
>> directly from viewers (consumers) by paid subscription. Instead, they
>> receive money from advertising based on the size of their viewer base. I
>> think that the MM team should figure out a way to get the funds from the
>> organizations that make lots of it and focus on increasing their user
>> base.
>
> There is no denying that it is hard to ask for money for something
> that has been free -- although some µManager users (and developers)
> have very kindly signed up for our service program early on; without
> this we would have been bankrupt already! (So thanks goes to those
> labs and companies.)
>
> But I think Sam's characterization (freemeum model) is a good way to
> think about this. It should be noted that the µManager user base is
> much larger (~10-fold?) than the subscribers of this mailing list.
> Presumably these users are finding µManager useful enough without
> getting support (at least not from us). And in the vast majority of
> cases, our service program pricing is very competitive with
> proprietary software, so I don't think this by itself should
> discourage use of µManager if we do our job and keep up with the
> development.
>
> I should also mention that, as the user base has gotten larger and as
> people have started doing diverse things with µManager, answering
> every single question that comes our way (e.g. via this mailing list)
> had already started to become unsustainable before we transitioned to
> Open Imaging. We would need a team of several engineers to handle all
> of those. We hope that the paid support model will allow us to scale
> our team with demand as more people subscribe.
>
>
>>> I certainly hope they would (and some of them do with more than just
>>> developing interfaces to their hardware, if that was all support there
>>> would still be a big gap to cover consisting of integration and GUI
>>> development amongst others), ...
>>
>> I was referring to direct financial support to Open Imaging by other
>> companies. i.e paying you to develop and integrate their devices with
>> micromanager. They could then use the resultant interfacing as a selling
>> incentive to end customers.
>
> This is something we might do as part of our business, but, as others
> have pointed out, µManager development is not just about adding device
> support. For this model to work on its own, we would have to charge a
> very high price for development of device support in order to
> subsidize development of the core program and user support.
>
>
> On Thu, Mar 10, 2016 at 4:48 PM, Dr.MVH <[hidden email]> wrote:
>> JonD wrote
>>> I'm pretty sure Open Imaging accepts money from companies to develop
>>> Micro-Manager device adapters.  Many companies (like mine) would prefer to
>>> develop device adapters themselves, in this way they achieve the same
>>> result via different means (i.e. they pay their employees to do the work
>>> instead of paying Open Imaging).
>>
>> Wouldn't that qualify as outsourcing Open Imaging at the expense of the end
>> user? On the premise that an in-house developed device adapter is of market
>> value to many companies (including yours) as you say, then the existence of
>> Open Imaging and MicroManager is automatically of value to you. Shouldn't
>> that be a reason to justify a little cut from your profits to Open Imaging,
>> depending whether the end user shows interest in using your technology with
>> MicroManager?
>
> We hope so, too, and for that reason we are also starting to offer
> services (even before development per se) and sponsorship
> opportunities to device manufacturers. This will become a little more
> visible once we have an official µManager release from Open Imaging
> (we're working very hard on this right now).
>
> On the other hand, we don't want to be "owned" solely by (a subset of)
> device manufacturers, either. I think the µManager community derives
> its strength from participation from both scientists/users and device
> manufacturers, and it makes sense for us to offer paid benefits to
> both of those groups.
>
> (Also, the "shouldn't that be a reason to justify a little cut"
> argument can be applied to well-funded scientists, too, if they are
> saving money by not buying proprietary software :)
>
>
>> JonD wrote
>>> Having (working) device adapters can definitely can be a selling point for
>>> hardware no matter how they get created.  But for some companies the
>>> marginal sales may not be worth the expense.
>>
>> Is there really a high expense involved in writing a device adapter for a
>> standardized USB3 technology, using standardized imaging sensors, in
>> standardized programming language when an experienced programmer has access
>> to the SDK provided by the vendor?
>
> Very roughly speaking, probably a few to several tens of thousands of
> dollars, if you'd like to know (camera adapters are the most complex).
> Of course it depends on many details. "Standard" interfaces are more
> work (to get right) than vendor-specific ones, because they need to
> work with a wider range of devices.
>
> Actually I've always hoped for µManager to have standard GenICam
> (USB3Vision/GigEVision) support, it's just that it's a relatively
> major project and never has been high enough on the list of
> high-priority tasks.
>
> I can imagine that for a company selling relatively inexpensive
> cameras (therefore in large quantities, and primarily outside of
> microscopy), additional sales created by µManager support would not be
> attractive enough.
>
>
>> JonD wrote
>>>  In which case the end user (e.g. you) are free to pay Open Imaging to
>>> develop a device adapter or else spend the time to create it yourself.  Of
>>> course many will just "vote with their feet" and buy hardware from
>>> companies that are providing and supporting device adapters for their
>>> hardware either directly or via Open Imaging's services.
>>>
>>> Jon
>>
>> If the end user had an option to purchase either a vendor's camera that
>> supports Micro-Manager for 4500 or an identical camera (i.e identical
>> sensor, albeit with different housing and stickers) from another vendor
>> without Micro-Manager support for 1000, which option would carry more
>> probability, assuming the end user is your average biologist that does a bit
>> of basic googling?  Perhaps a lot (vast majority) of users will prefer the
>> 1000 camera and would go via DirectShow and OpenCVgrabber, even if the
>> controls are not making full use of the cameras capabilities but would
>> suffice for certain experimental set ups.
>
> Maybe. It depends on what 'typical' is (a dangerous word to apply to
> scientists). µManager was originally developed in a lab that does
> serious microscopy, and initially adopted in similar labs (think:
> cameras costing $10k-50k). It's great that we have gradually also come
> to support less expensive cameras, vastly expanding our potential user
> base, but (as I have previously noted on this mailing list) the
> question of who should pay for development of support for the least
> expensive range of cameras remains a hard one, because in most cases
> nobody makes a huge amount of money from it.
>
> In fact, we (Open Imaging) are trying to be the entity that works on
> the important aspects of µManager that otherwise nobody would work on.
> Support for standard camera interfaces is the sort of thing that we
> might want to tackle ourselves maybe if we get enough subscriptions in
> the future to expand our team.
>
> In the short term, with our limited resources, we will likely need to
> concentrate on higher-priority tasks in the core program, unless
> somebody wants to specifically pay us to work on it, unfortunately.
>
> Best regards,
> Mark
>
>
> Mark A. Tsuchida
> CEO & Co-CTO
> Open Imaging, Inc.
> https://open-imaging.com
>
> ------------------------------------------------------------------------------
> Transform Data into Opportunity.
> Accelerate data analysis in your applications with
> Intel Data Analytics Acceleration Library.
> Click to learn more.
> http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
> _______________________________________________
> micro-manager-general mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/micro-manager-general


------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785111&iu=/4140
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Dr.MVH
In reply to this post by Mark Tsuchida-3
Dear all,
Since I am not a programmer, I find it hard quoting parts of the several previous replies in a nested fashion. I therefore encourage new and seasoned members of the community to look at the "List" view of this thread on Nabble for the best semantic/contextual view (and my favorite) and to PLEASE READ the posts (replies) by Nico Stuurman and Mark Tsuchida because it´s worth it. I hope the forum's etiquette will be understanding. Your replies (Nico, Mark and others) are testimony to the community's commitment to an open source tool for scientists. My "devils advocate" remarks were not ill-intended (90% of the time are not).

I started this newbie post looking for a practical solution to the problem described in the title of the subject. I discovered Micro-Manager in December 2015 (where was I living??). I obviously failed to interface the Point Grey camera USB3 vision with Micro-Manager (MM), although i managed to interface a few old devices with MM and used a webcam to test the accuracy (100% accuracy). A scientific-grade camera was the last part I needed to complete the system and "degrade" sales representatives that attempt to come across as "holier than thou". At the same time I wanted to highlight the merits of MM to the community here in the University of Nurnberg/Erlangen in Germany.
When asked by the department about live microscopy at the various facilities I will give them three options:
1. Another camera of inferior specifications valued as the Grasshopper3 but interfaces to MM
2. A camera equivalent to the Grasshopper 3 that interfaces to MM but costs 5 x more
3. A new software that probably costs 8 x more.
My recommendation will be number 1.
However if someone from Open Imaging says that they can interface the Grasshopper 3 USB3 with MM for the base subscription plus two "technical consultations" (something around 1500$), I will surely tell the institute to go for that. When it comes to my own grants, I would not hesitate signing for a subscription to Open Imaging, since microscopy (the MM sort) is integral to my research interests.

And now to the business(community)-oriented stuff. I mostly agree with all the explanations (about many aspects, not only device adapters as it was pointed out) given to my opinions concerning the Open Imaging business model. I stand corrected.

Indeed, youtube has a larger user base and a comparison between youtube creators and MM developers is naive, initially. But how many fold larger is a youtube fan base compared to the MM user base?  According to Mark the MM user base is  "(~10-fold?) than the subscribers of this mailing list ". So, 1048 x 10 = 10,480 users. A VERY VERY profitable channel (creator) on youtube has 1 million (1,000,000) views a day (and depending on the engagement) making on average 2$ per 1000 views which translates to 730.000 per year. If every MM user contributed 100$ a year to Open Imaging that would translate to
~1 million $ per year (much better than 400,000$ / year from NIH?). This figure could also be reached if 10% of MM users gave 1000$ a year for MM. It would surpass a very engaging youtube content creator channel (usually childish individuals with no programming experience). Thus, the market dynamics between a Youtube channel's subscribers and Micro-Manager users (subscribers) are not as different as it first appears? Bear in mind that Micro-Manager is very engaging!!

Finally,
Mark said
" In the short term, with our limited resources, we will likely need to
concentrate on higher-priority tasks in the core program, unless
somebody wants to specifically pay us to work on it, unfortunately. "

I hope Open Imaging is not absorbed by a larger company willing to splash 5-10 million in an attempt to avoid collateral damage in the long term. That would be a pity.

Any help to my little initial problem BTW will be welcomed and appreciated.

Best Regards
Michel

 







Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

JonD
Administrator
Dr.MVH wrote
Since I am not a programmer, I find it hard quoting parts of the several previous replies in a nested fashion.
It can be a pain.  I have come to appreciate the Nabble archive interface and then you just manually copy/paste the <⁄quote> tag between sections to quote.

Dr.MVH wrote
Finally, Mark said
" In the short term, with our limited resources, we will likely need to
concentrate on higher-priority tasks in the core program, unless
somebody wants to specifically pay us to work on it, unfortunately. "

I hope Open Imaging is not absorbed by a larger company willing to splash 5-10 million in an attempt to avoid collateral damage in the long term. That would be a pity.
A large company could in theory purchase Open Imaging and/or hire away Mark and Chris (or other major contributors to MM).  But no company can defensively stop MM from being distributed for free nor can they prevent others from improving MM.  That is the powerful nature of open-source software.  In fact, Open Imaging could eventually have "competitors" whose business is also supporting/furthering MM.

The worst-case scenario is that MM stagnates because nobody is developing it further.  There are plenty of users (and companies) hoping to avoid that fate.  Hopefully they are "putting their money where their mouth is" and supporting Open Imaging via subscriptions (and corporate sponsorships) and/or are directly contributing to the MM source code.

Jon
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Mark Tsuchida-3
In reply to this post by Dr.MVH
Hi Michel,

On Fri, Mar 11, 2016 at 9:06 PM, Dr.MVH <[hidden email]> wrote:
> I hope Open Imaging is not absorbed by a larger company willing to splash
> 5-10 million in an attempt to avoid collateral damage in the long term. That
> would be a pity.

That's certainly not something we're aiming or planning for. Speaking
for myself (as one of the cofounders), the goal is to perpetuate
Micro-Manager and continue to innovate, and Open Imaging is a means to
that end. It's possible that other ways to achieve that goal may arise
in the future (though none are in sight), so I won't make any specific
promises about future directions. But selling out to a (hypothetical)
company that just wants to shut down Micro-Manager (or otherwise
cripple its openness and neutrality) would very much go against what
I'm in this for.

Best,
Mark


Mark A. Tsuchida
CEO & Co-CTO
Open Imaging, Inc.
https://open-imaging.com

------------------------------------------------------------------------------
Transform Data into Opportunity.
Accelerate data analysis in your applications with
Intel Data Analytics Acceleration Library.
Click to learn more.
http://pubads.g.doubleclick.net/gampad/clk?id=278785231&iu=/4140
_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

sapshuk
In reply to this post by Dr.MVH
Dr.MVH wrote
> Dr.MVH wrote
>> Does anybody know whether the GigE device adapter was extended (or can be
>> easily extended) to make use of the new JAI SDK (2.1.6) version? That
>> would theoretically provide interfacing to Micro-manager for a lot of the
>> cameras that use the USB3 Vision standard. Since I am not a programmer I
>> do not know how easy it is to recompile the GigE device adapter to make
>> use of the new JAI SDK.

>> Michel


> Really, what we need to do is get Point Grey to recognize the value in
> writing device adapters for Micro-Manager. I'm mentioned it to them before,
> but I should do it again. And anyone else interested should do the same.
>
> -Sam

Do you want to put together an open letter to Point Grey? I would be
happy to add my name.

Kurt

_______________________________________________
micro-manager-general mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/micro-manager-general

Dear Sam,
Dear Kurt,
I have already contacted Point Grey and their answer was that "gathering from their customer feedback their cameras are not compatible with this tool (MicroManager) and therefore support can not be extended..." I would be very interested in convincing them to compile a MicroManager device adapter for their cameras. However, I also think that a general MicroManager device adapter for USB3-Vision cameras should exist.  I am currently in contact with Point Grey, looking into some alternative ways. I will be keeping this thread up to date with the developments as to help other interested individuals in  PGR-MicroManager interfacing. Let´s wait till the end of the week for more parties willing to join an open letter effort to Point Grey for a MicroManager device adapter.

As a side note, I have my reservations about the new business model of the MicroManager developing team at Open Imaging and the subscription-based model. Perhaps companies like Point Grey and others that profit mainly from camera and equipment sales should support Open Imaging in developing interfacing with their products. Unfortunately, I noticed that these companies are often interested in business partners that offer closed source software for their products, apparently for increasing profits. Your feedback is very welcome.

Michel


Reverting back to the original discussion:

Did you end up contacting Point Grey about this, Michel? I have 2 Grasshopper USB 3 cameras which I tried to operate and synchronise using MicroManager, without any success. This includes using the JAI SDK and attempts to write a device adapter as well. I finally ended up using an arduino board for doing this, since their FlyCapture software is not nearly as versatile or straightforward as Micromanager anyway. I wonder how they concluded from their customer feedback that an adapter for uM will not be interesting, when I have been bugging them about this for over a year now! Their support team was quite helpful with the synchronization of cameras, but they have alwas been very evasive about device adapters for uM. I would be therefore very glad to join in the open letter effort.  

Cheers,
Sapna
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: point grey research (PGR) Grasshopper 3 USB3 vision interfacing with MM

Florian Fahrbach
So this discussion went very far off topic.
This is my first encounter with micro manager. Microscope and Andor Camera are working fine, but I also want a USB 3 vision camera in micro manager.
Has any option become available in the mean time?

Thanks
Flo
12
Loading...