Anyone compared recently released sCMOS? (PCO 4.2 / Zyla 4.2 / Flash4 V2)?

[Emmanuel Levy]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Dear All,

We intend to buy a spinning disk confocal microscope and we are therefore
looking for the best camera to suit our need.

The main use will be imaging of fluorescently tagged proteins expressed in
live cells.

One thing is certain, we will buy a sCMOS, but we are unsure of which one
is best.

For a while, it is my understanding that the Hamamatsu Flash4 V2 was the
best - but couple of month ago the same chip has been used in the Zyla 4.2
(Andor) and in the PCO Edge 4.2.

All of them exhibit highly similar characteristics, with QE ~70%, low read
noise, 100fps with full chip, etc.

I guess the only way to see what exactly are the advantages/drawbacks would
be to use them in the same setup and compare them side by side.
Unfortunately I cannot do that and therefore I'd be happy to hear about any
experience you have.

I insist that only the late models should be compared (released in the fall
of 2013 I believe). If you have had a bad experience with previous models
(e.g., Andor Neo), it does not apply here.

Thanks in advance,

Emmanuel

[Andrew York]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

 I use the pco edge 5.5 and the pco edge 4.2, and other members of my lab
use the Hamamatsu 4.2. They're all great cameras. The major difference for
me between the pco and the Hamamatsu is that the pco SDK is freely
available:
http://www.pco.de/fileadmin/user_upload/db/download/MA_DCSDKWINE_114.pdf
which allows me to write and publish code that talks to the pco camera.

 The Hamamatsu SDK costs extra, and prohibits us from sharing code we write
that talks to Hamamatsu cameras. This was a dealbreaker for me, so we
complained to Hamamatsu. They were pretty cool about it, listened to our
concerns, and worked out an acceptable solution. However, this process took
too long and I didn't want to delay my project, so I'm using a pco 4.2 now.
Other members of my lab had already bought Hamamatsu cameras and they're
stuck using labview to control them.

 My understanding is that the lower QE of the 5.5 chip is (part of) a
feature, not a bug: you can have 5 transistors, lower QE, and choice of
'global' or 'rolling' shutter, or you can have 4 transistors, higher QE,
and only 'rolling' shutter. Rolling shutter operation requires you to use
pulsed, synchronized illumination, but is also lower noise, so we always
use rolling shutter on the 5.5. If you don't use pulsed illumination, you
get an effective 'blurring' in time across your chip. Since the whole point
of an sCMOS is to go fast, it seems silly to me to use continuous
illumination with rolling shutter in most cases. Of course, if you do use
pulsed illumination, your illumination duty cycle goes down (to zero at top
speed!)

I have no experience with Andor's sCMOS, and would be curious to hear how
they compare. Cooling? Does that matter at high speeds? Kurt Thorn has
written about them:
http://nic.ucsf.edu/blog/?p=464
I'm not sure if he's used comparable pco or Hamamatsu versions.


On Sat, Apr 19, 2014 at 12:22 PM, Emmanuel Levy <[hidden email]>wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Dear All,
>
> We intend to buy a spinning disk confocal microscope and we are therefore
> looking for the best camera to suit our need.
>
> The main use will be imaging of fluorescently tagged proteins expressed in
> live cells.
>
> One thing is certain, we will buy a sCMOS, but we are unsure of which one
> is best.
>
> For a while, it is my understanding that the Hamamatsu Flash4 V2 was the
> best - but couple of month ago the same chip has been used in the Zyla 4.2
> (Andor) and in the PCO Edge 4.2.
>
> All of them exhibit highly similar characteristics, with QE ~70%, low read
> noise, 100fps with full chip, etc.
>
> I guess the only way to see what exactly are the advantages/drawbacks would
> be to use them in the same setup and compare them side by side.
> Unfortunately I cannot do that and therefore I'd be happy to hear about any
> experience you have.
>
> I insist that only the late models should be compared (released in the fall
> of 2013 I believe). If you have had a bad experience with previous models
> (e.g., Andor Neo), it does not apply here.
>
> Thanks in advance,
>
> Emmanuel
>

[Kurt Thorn]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

On 4/20/2014 8:41 AM, Andrew York wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
>   I use the pco edge 5.5 and the pco edge 4.2, and other members of my lab
> use the Hamamatsu 4.2. They're all great cameras. The major difference for
> me between the pco and the Hamamatsu is that the pco SDK is freely
> available:
> http://www.pco.de/fileadmin/user_upload/db/download/MA_DCSDKWINE_114.pdf
> which allows me to write and publish code that talks to the pco camera.
>
>   The Hamamatsu SDK costs extra, and prohibits us from sharing code we write
> that talks to Hamamatsu cameras. This was a dealbreaker for me, so we
> complained to Hamamatsu. They were pretty cool about it, listened to our
> concerns, and worked out an acceptable solution. However, this process took
> too long and I didn't want to delay my project, so I'm using a pco 4.2 now.
> Other members of my lab had already bought Hamamatsu cameras and they're
> stuck using labview to control them.
>
>   My understanding is that the lower QE of the 5.5 chip is (part of) a
> feature, not a bug: you can have 5 transistors, lower QE, and choice of
> 'global' or 'rolling' shutter, or you can have 4 transistors, higher QE,
> and only 'rolling' shutter. Rolling shutter operation requires you to use
> pulsed, synchronized illumination, but is also lower noise, so we always
> use rolling shutter on the 5.5. If you don't use pulsed illumination, you
> get an effective 'blurring' in time across your chip. Since the whole point
> of an sCMOS is to go fast, it seems silly to me to use continuous
> illumination with rolling shutter in most cases. Of course, if you do use
> pulsed illumination, your illumination duty cycle goes down (to zero at top
> speed!)
>
> I have no experience with Andor's sCMOS, and would be curious to hear how
> they compare. Cooling? Does that matter at high speeds? Kurt Thorn has
> written about them:
> http://nic.ucsf.edu/blog/?p=464
> I'm not sure if he's used comparable pco or Hamamatsu versions.

We have two Flash 4.0 cameras, one of which is currently attached to a
microscope (the other one should be set up next week; we had some
trouble getting a PC set up to handle the camera and some of the legacy
hardware on that microscope). Both the Flash 4.0 and the Zyla 5.5 work
well. I have not done detailed side-by-side comparisons of the Zyla 5.5
and the Flash 4.0 yet.  I hope to also get a PCO 4.2 this summer and
test all three.

One thing I will say is that, in my opinion, the extra field of view on
the 5.5 MP chip is not all that useful. We see some vignetting and
significant aberrations at the edges of the field of view, so we almost
always run our Zyla cropped to 2k x 2k.  My sense, without detailed
testing, is that the Zyla and Flash 4.0 are pretty similar and that you
would only notice the difference between them if you are really pushing
the limits of the camera. We also mostly use the Zyla in rolling shutter
mode with pulsed illumination, so the global shutter is not that
important to us.

Kurt

>

[Neil Anthony]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Emmanuel, I'm not sure if it's one of the latest options or not, but has
anybody looked at the optiMOS from QImaging?

It has a CIS1920F sCMOS sensor, and everything else looks as good as the
others but with a reduced sensor size (1920x1080).  It's less that 10k,
so it's considerably cheaper.  They don't mention the shutter type, but
it has to be a rolling shutter, right?

Also, there's the Flash 4.0 LT.  I think it's closer to 12k instead of
the 16-20k of the fastest and global shutter versions.  It's USB 3.0
which makes life a little easier, and it's got a better QE and less
noise than the cheaper version of the Zyla 5.5 (which does offer 30%
more real estate; 2561x2160 vs. 2048x2048 both with 6.5µm pixels).

Does anybody know of a webpage/document out there that has all the
details and prices in one place?  It would be great to condense all
these things.

Neil


On 4/19/2014 12:22 PM, Emmanuel Levy wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Dear All,
>
> We intend to buy a spinning disk confocal microscope and we are therefore
> looking for the best camera to suit our need.
>
> The main use will be imaging of fluorescently tagged proteins expressed in
> live cells.
>
> One thing is certain, we will buy a sCMOS, but we are unsure of which one
> is best.
>
> For a while, it is my understanding that the Hamamatsu Flash4 V2 was the
> best - but couple of month ago the same chip has been used in the Zyla 4.2
> (Andor) and in the PCO Edge 4.2.
>
> All of them exhibit highly similar characteristics, with QE ~70%, low read
> noise, 100fps with full chip, etc.
>
> I guess the only way to see what exactly are the advantages/drawbacks would
> be to use them in the same setup and compare them side by side.
> Unfortunately I cannot do that and therefore I'd be happy to hear about any
> experience you have.
>
> I insist that only the late models should be compared (released in the fall
> of 2013 I believe). If you have had a bad experience with previous models
> (e.g., Andor Neo), it does not apply here.
>
> Thanks in advance,
>
> Emmanuel
>

[Loralei Dewe-3]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Neil,

The optiMOS is a very nice camera, it does have the rolling shutter, QE is
about 55%  but it still runs the 100 fps at full res and is compatible with
many of the current software packages ;-)

The sCMOS cameras that use the USB 3.0 are not going to run as fast with
the USB restriction. The pco cameras run at 40 fps but I believe the
Hamamatsu cameras only do 30 fps. Correct me if I am mistaken..

Loralei


On Fri, May 16, 2014 at 7:26 AM, Neil Anthony <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Emmanuel, I'm not sure if it's one of the latest options or not, but has
> anybody looked at the optiMOS from QImaging?
>
> It has a CIS1920F sCMOS sensor, and everything else looks as good as the
> others but with a reduced sensor size (1920x1080).  It's less that 10k, so
> it's considerably cheaper.  They don't mention the shutter type, but it has
> to be a rolling shutter, right?
>
> Also, there's the Flash 4.0 LT.  I think it's closer to 12k instead of the
> 16-20k of the fastest and global shutter versions.  It's USB 3.0 which
> makes life a little easier, and it's got a better QE and less noise than
> the cheaper version of the Zyla 5.5 (which does offer 30% more real estate;
> 2561x2160 vs. 2048x2048 both with 6.5µm pixels).
>
> Does anybody know of a webpage/document out there that has all the details
> and prices in one place?  It would be great to condense all these things.
>
> Neil
>
>
>
> On 4/19/2014 12:22 PM, Emmanuel Levy wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> Post images on http://www.imgur.com and include the link in your posting.
>> *****
>>
>> Dear All,
>>
>> We intend to buy a spinning disk confocal microscope and we are therefore
>> looking for the best camera to suit our need.
>>
>> The main use will be imaging of fluorescently tagged proteins expressed in
>> live cells.
>>
>> One thing is certain, we will buy a sCMOS, but we are unsure of which one
>> is best.
>>
>> For a while, it is my understanding that the Hamamatsu Flash4 V2 was the
>> best - but couple of month ago the same chip has been used in the Zyla 4.2
>> (Andor) and in the PCO Edge 4.2.
>>
>> All of them exhibit highly similar characteristics, with QE ~70%, low read
>> noise, 100fps with full chip, etc.
>>
>> I guess the only way to see what exactly are the advantages/drawbacks
>> would
>> be to use them in the same setup and compare them side by side.
>> Unfortunately I cannot do that and therefore I'd be happy to hear about
>> any
>> experience you have.
>>
>> I insist that only the late models should be compared (released in the
>> fall
>> of 2013 I believe). If you have had a bad experience with previous models
>> (e.g., Andor Neo), it does not apply here.
>>
>> Thanks in advance,
>>
>> Emmanuel
>>
>>

[Gerhard Holst]

Hi Loralei,

I would like to know if the optiMOS is really capable of transferring the 1920 x 1080 pixel 16 bit data via USB 3.0. If I am not mistaken, then we have (1920 x 1080 pixel) x 2 Byte (16 Bit Dynamic) x 100 frames/s = 414.7 MByte/s, which is above the theoretical limit of 400 MByte/s of available hardware parts used for USB 3.0. For practical applications the limit is even lower 340 - 360 MByte/s.
So maybe the data are stored in the camera and transferred slower or compressed. And yes, the USB 3.0 cameras with the larger sCMOS image sensors, which are usually double as fast compared to the small sensor (if not operated in Time delay integration or Lightsheet mode), they are limited by the data transfer bandwidth of the USB 3.0. Therefore some manufacturers have decided just to offer the slower performance of the sensor.

with best regards,

Gerhard
_______________________________

Dr. Gerhard Holst
Science & Research
PCO AG
Donaupark 11
93309 Kelheim, Germany
fon [phone]
fax [phone]
mob [phone]
[hidden email]
www.pco.de


-----Ursprüngliche Nachricht-----
Von: Confocal Microscopy List [mailto:[hidden email]] Im Auftrag von Loralei Dewe
Gesendet: Freitag, 16. Mai 2014 21:22
An: [hidden email]
Betreff: Re: Anyone compared recently released sCMOS? (PCO 4.2 / Zyla 4.2 / Flash4 V2)?

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Neil,

The optiMOS is a very nice camera, it does have the rolling shutter, QE is
about 55%  but it still runs the 100 fps at full res and is compatible with
many of the current software packages ;-)

The sCMOS cameras that use the USB 3.0 are not going to run as fast with
the USB restriction. The pco cameras run at 40 fps but I believe the
Hamamatsu cameras only do 30 fps. Correct me if I am mistaken..

Loralei


On Fri, May 16, 2014 at 7:26 AM, Neil Anthony <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Emmanuel, I'm not sure if it's one of the latest options or not, but has
> anybody looked at the optiMOS from QImaging?
>
> It has a CIS1920F sCMOS sensor, and everything else looks as good as the
> others but with a reduced sensor size (1920x1080).  It's less that 10k, so
> it's considerably cheaper.  They don't mention the shutter type, but it has
> to be a rolling shutter, right?
>
> Also, there's the Flash 4.0 LT.  I think it's closer to 12k instead of the
> 16-20k of the fastest and global shutter versions.  It's USB 3.0 which
> makes life a little easier, and it's got a better QE and less noise than
> the cheaper version of the Zyla 5.5 (which does offer 30% more real estate;
> 2561x2160 vs. 2048x2048 both with 6.5µm pixels).
>
> Does anybody know of a webpage/document out there that has all the details
> and prices in one place?  It would be great to condense all these things.
>
> Neil
>
>
>
> On 4/19/2014 12:22 PM, Emmanuel Levy wrote:
>
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> Post images on http://www.imgur.com and include the link in your posting.
>> *****
>>
>> Dear All,
>>
>> We intend to buy a spinning disk confocal microscope and we are therefore
>> looking for the best camera to suit our need.
>>
>> The main use will be imaging of fluorescently tagged proteins expressed in
>> live cells.
>>
>> One thing is certain, we will buy a sCMOS, but we are unsure of which one
>> is best.
>>
>> For a while, it is my understanding that the Hamamatsu Flash4 V2 was the
>> best - but couple of month ago the same chip has been used in the Zyla 4.2
>> (Andor) and in the PCO Edge 4.2.
>>
>> All of them exhibit highly similar characteristics, with QE ~70%, low read
>> noise, 100fps with full chip, etc.
>>
>> I guess the only way to see what exactly are the advantages/drawbacks
>> would
>> be to use them in the same setup and compare them side by side.
>> Unfortunately I cannot do that and therefore I'd be happy to hear about
>> any
>> experience you have.
>>
>> I insist that only the late models should be compared (released in the
>> fall
>> of 2013 I believe). If you have had a bad experience with previous models
>> (e.g., Andor Neo), it does not apply here.
>>
>> Thanks in advance,
>>
>> Emmanuel
>>
>>

[Stanislav Vitha-2]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

I just looked at the optiMOS specs, it uses a proprietary camera interface,
not USB3. The company also recommends to use a solid state disk for
storage, in order to sustain the data transfer rate.


Stan Vitha
Microscopy and Imaging Center
Texas A&M University



=================================
Subject: AW: Anyone compared recently released sCMOS? (PCO 4.2 / Zyla
4.2 / Flash4 V2)?
From: Gerhard Holst <[hidden email]>
Reply-To: Confocal Microscopy List
<[hidden email]>
Date: Mon, 19 May 2014 09:07:04 +0200
Content-Type: text/plain


Hi Loralei,
 
I would like to know if the optiMOS is really capable of transferring the
1920 x 1080 pixel 16 bit data via USB 3.0. If I am not mistaken, then we
 
with best regards,
 
Gerhard

 

[Loralei Dewe-3]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Gerhard,

I never said the optiMOS uses a USB 3.0 interface. It does not, it uses a
camera link card similar to the pco.edge.  What I did say was that the
sCMOS that do use a USB 3.0 connection will not be able to run faster than
30-40 fps.

Cheers,
Loralei


On Tue, May 20, 2014 at 7:18 AM, Stanislav Vitha <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> I just looked at the optiMOS specs, it uses a proprietary camera interface,
> not USB3. The company also recommends to use a solid state disk for
> storage, in order to sustain the data transfer rate.
>
>
> Stan Vitha
> Microscopy and Imaging Center
> Texas A&M University
>
>
>
> =================================
> Subject: AW: Anyone compared recently released sCMOS? (PCO 4.2 / Zyla
> 4.2 / Flash4 V2)?
> From: Gerhard Holst <[hidden email]>
> Reply-To: Confocal Microscopy List
> <[hidden email]>
> Date: Mon, 19 May 2014 09:07:04 +0200
> Content-Type: text/plain
>
>
> Hi Loralei,
>
> I would like to know if the optiMOS is really capable of transferring the
> 1920 x 1080 pixel 16 bit data via USB 3.0. If I am not mistaken, then we
>
> with best regards,
>
> Gerhard
>
>
>

[Gerhard Holst]

Hi Loralei,

sorry, my misunderstanding. Yes, it is correct that the USB 3.0 limits the data transfer rate, therefore there are the Camera Link sCMOS camera models as well. If then the smaller chip is compared to the larger ones, one have to know that the smaller one corresponds to one half of the larger chips, which means it  has one rolling shutter instead of the two of the larger ones, which is the reason why the larger sensors, if a ROI with the same resolution as the smaller one is used, are always double as fast. But I guess you might know that already.


with best regards,
Gerhard
_______________________________

Dr. Gerhard Holst
Science & Research
PCO AG
Donaupark 11
93309 Kelheim, Germany
fon [phone]
fax [phone]
mob [phone]
[hidden email]
www.pco.de


-----Ursprüngliche Nachricht-----
Von: Confocal Microscopy List [mailto:[hidden email]] Im Auftrag von Loralei Dewe
Gesendet: Dienstag, 20. Mai 2014 20:19
An: [hidden email]
Betreff: Re: AW: Anyone compared recently released sCMOS? (PCO 4.2 / Zyla 4.2 / Flash4 V2)?

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hi Gerhard,

I never said the optiMOS uses a USB 3.0 interface. It does not, it uses a
camera link card similar to the pco.edge.  What I did say was that the
sCMOS that do use a USB 3.0 connection will not be able to run faster than
30-40 fps.

Cheers,
Loralei


On Tue, May 20, 2014 at 7:18 AM, Stanislav Vitha <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> I just looked at the optiMOS specs, it uses a proprietary camera interface,
> not USB3. The company also recommends to use a solid state disk for
> storage, in order to sustain the data transfer rate.
>
>
> Stan Vitha
> Microscopy and Imaging Center
> Texas A&M University
>
>
>
> =================================
> Subject: AW: Anyone compared recently released sCMOS? (PCO 4.2 / Zyla
> 4.2 / Flash4 V2)?
> From: Gerhard Holst <[hidden email]>
> Reply-To: Confocal Microscopy List
> <[hidden email]>
> Date: Mon, 19 May 2014 09:07:04 +0200
> Content-Type: text/plain
>
>
> Hi Loralei,
>
> I would like to know if the optiMOS is really capable of transferring the
> 1920 x 1080 pixel 16 bit data via USB 3.0. If I am not mistaken, then we
>
> with best regards,
>
> Gerhard
>
>
>