*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Another thing to think about is exactly what image processing type/image processing tool you will use. As an example, we routinely use Matlab for custom image processing for our clients.  Execution speeds are a routine consideration for the larger - many gigapixel - images we get from our whole slide images. Matlab does a nice job of using both the CPU and GPU available on the system. Optimizing the CPU is probably to be expected at this point but taking advantage of the GPU can give significant improvements in processing speed. We routinely see differences in execution times in the 1X to 10X range. 

This is all a long winded way of saying that this only works with graphics that support Nvidia's GPU parallel computing platform called CUDA. Here are couple links (I have no commercial affiliation with either company) 
What is CUDA | NVIDIA Developer Zone.  
MATLAB GPU Computing with NVIDIA CUDA-Enabled GPUs

There are myriad software set-ups of course but you should definitely think about matching your software with hardware if there are large data sets to be processed. A very similar price might end up with very different day to day processing times. 

Good luck and best wishes!

Casey 



-----------------------------------------------
Casey Laris 
Reveal Biosciences


On Mar 13, 2013, at 3:20 AM, "Scott, Mark" <[log in to unmask]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
> 
> Dear Arvydas,
> 
> Obviously a lot will depend on the software packages you plan to run, but in my experience Linux seems to cope better with heavy processing tasks on large data sets than Windows traditionally does, i'm sure someone with more knowledge of the inner workings of this may be able to shed light on it but it seems to me that  Ubuntu (and Suse which i've used in the past) utilise multiple cores a lot better than Windows but perhaps i'm just stretching.
> 
> The best thing to do would be to dual boot a PC and have both Windows AND Ubuntu running and test them out for your needs.   I would be very keen to hear how the results go since i'm only running Ubuntu under VMWare rather than dual booting it.
> 
> Thanks
> Mark
> 
> 
> 
> -----Original Message-----
> From: Confocal Microscopy List [mailto:[log in to unmask]] On Behalf Of Arvydas Matiukas
> Sent: 12 March 2013 17:53
> To: [log in to unmask]
> Subject: Re: Computer for image analysis...OS choice
> 
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
> 
> Dear listers/image analysts,
> 
> Our discussion so far mostly concerned hardware and to some extent
> image analysis software. Just few mentioning of Mac vs PC.
> 
> I am a PC person which still gives few OS choices: Windows (7 Pro x64,
> 7 Ultimate x64), Linux (Ubuntu?), ...BSD?
> I wonder how the choice of OS may affect image analysis software 
> performance and manipulation of extra large files (>10GB  see Cameron
> posting below).
> 
> Best,
> Arvydas
> 
> 
>>>> Cameron Nowell <[log in to unmask]> 3/11/2013 4:31 PM >>>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
> 
> Hi All,
> 
> I thought I might chip in one of our recent experiences. Firstly all the discussions going on re components I fully agree with. One thing that has been mentioned is gaming cards. For sure go for the best gaming card you can lay your hands on (even the most expensive are not that much for a total beast of a card). The issue with buying pre-configured workstations is that they will all come with a Nvidia Quadro or AMD FireGL card. These are workstation glass cards that come with (supposably) better support and refined drivers. They also carry a price premium of 2-3x on top of the equivalent gaming card. So if you can build your own system or buy a preconfigured workstation with the lowest end card you can and upgrade it yourself. Just make sure the power supply can take it.
> 
> Now one thing from our recent experience is file formats. We have a user who collects huge, and I mean really huge, data sets. These are between 50 and 120GB for each file. Large area tile scans, 4-5 channels and 200+ z slices. Yeah scary. Now if these are captured on a leica system there doesn't seem to be any problem dealing with them (opening, stitching, manipulating etc). but when they are capture on a zeiss system the LSM file format seems to bring things to a grinding halt. There doesn't seem to be the capability to open parts of the image as needed, the whole thing has to get loaded into ram (and if there isn't enough ram it then gets paged). Interestingly if you convert it all to say an imaris ims file, it uses vaery minimal resources and can be easily manipulated.
> 
> So even if you have a true beast of a machine, different file formats may bring it to its knees anyway.
> 
> Cheers
> 
> Cam
> 
> 
> Cameron J. Nowell
> Centre for Dynamic Imaging
> The Walter and Eliza Hall Institute of Medical Research
> 1G Royal Parade
> Parkville, Victoria 3052
> Australia
> 
> Phone: +61 3 9345 2871
> Mobile: +61422882700
> Fax: +61 3 9347 0852
> 
> Facility Website
> LinkedIn Profile
> 
> 
> 
> -----Original Message-----
> From: Confocal Microscopy List [mailto:[log in to unmask]] On Behalf Of Arvydas Matiukas
> Sent: Saturday, 9 March 2013 4:24 AM
> To: [log in to unmask]
> Subject: Computer for image analysis
> 
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
> 
> Dear listers/microscopists,
> 
> I assume there is good time to update new trends in image analysis hardware. The last discussions on image analysis computer were in 2006-8. Though the basic principles of CPU, RAM, hard drive, video card, monitor selection still hold some new types of hardware became popular/available, e.g. SSD drives, APU, water cooling.
> Now a decent gaming computer (~$1k) has the processing power of a 2006 expensive workstation (~$20K). I was suprised that I was able to completely overhaul my 8 year old ATX case to a quad core 2GHz APU, 8GB 1600MHz RAM, 160GB SATA-2 SSD, water cooling, USB3 and SATA3 Gigabyte motherboard, and 4 monitor 1GB video card.
> for under $300 (online, after rebates).
> 
> Now I am wiling to upgrade/overhaul my work computer which is used to run ImageJ, Fiji, Deconvolution (Autoquant, Huygens), Matlab, PV-Vawe, Labview, Origin. Please advice/share you thoughts what best configuration is possible to buy for $2-3k (monitor excluded).
> My first choice would be  to go with a fast gaming computer, e.g.
> Dell-Alienware Aurora
> Windows* 7 Ultimate, 64Bit, English
> 2nd Generation Intel* Core* i7-3820 (10M Cache, Overclocked up to 4.1
> GHz)
> 16GB (4 X 4GB) Quad Channel DDR3 at 1600MHz
> NVIDIA* GeForce* GTX 660 1.5GB GDDR5
> 1TB RAID 0 (2x 500GB SATA 6Gb/s) Solid State Hybrid
> 19-in-1 Media Card Reader
> No Monitor
> Integrated 7.1 Channel Audio
> 
> The second  choice would be to buy all components online and build a computer myself (I have done this about 50 times over
> 25 years). This option typically saves money or buys better components, and provides you full specs of the hardware. The con of this approach is that it wastes some of your time to debug/make all the hardware work together and with your software. However, as the computer is for me not just a box but a tool I am ready to make this sacrifice.
> 
> BTW, is there any solid preference towards CPU Type (Intel ix/AMD/Intel
> Xeon)
> 
> Thanks for your input/advice/thoughts,
> Arvydas
> --------------------
> 
> 
> 
> 
> Arvydas Matiukas, Ph.D.
> Director of Confocal&Two-Photon Core
> Department of Neurosci& Physiology
> SUNY Upstate Medical University
> 766 Irving Ave., WH 3167
> Syracuse, NY 13210
> tel.: 315-464-7997
> fax: 315-464-8014
> email: [log in to unmask]
>>>> Tim Feinstein <[log in to unmask]> 3/7/2013 5:13 PM >>>
> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
> 
> Hi Chris, 
> 
> Agreed about FRET with a dichroic-based beam splitter on the emission side such as a DualView or QuadView.  No commercial interest, but our group uses a DV2 widefield/TIRF system for quantitative FRET just about all day every day.  As the executives say, drive for show and putt for dough.  In general I'd always go widefield if quantitative is more important than spatial.  
> 
> JP, unless I remember wrong the 510 has its detectors in a series with switchable dichroic filters between each detector (for example, a 515-ish nm longpass dichroic between detectors 1 and 2 if you want to see cerulean/venus/FRET).  Thus it collects each emission channel simultaneously and can go as fast as it can scan a line and alternate lasers, which pass through a two-notch excitation filter.  
> Creulean-FRET bleed-through is still a bear of course, but it always is, and there are long established ways to correct that.  
> 
> cheers, 
> 
> 
> TF
> 
> Timothy Feinstein, PhD
> Visiting Research Associate
> Laboratory for GPCR Biology
> Dept. of Pharmacology & Chemical Biology University of Pittsburgh, School of Medicine BST W1301, 200 Lothrop St.
> Pittsburgh, PA  15261
> 
> On Mar 7, 2013, at 4:23 PM, Chris Tully wrote:
> 
>> *****
>> To join, leave or search the confocal microscopy listserv, go to:
>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>> *****
>> 
>> If speed is the issue and you can illuminate with broad spectrum
> light, you
>> might want to consider the QV2 from Photometrics:
>> http://www.photometrics.com/products/multichannel/
>> 
>> It allows a single camera to simultaneously image up to 4 channels
> by
>> splitting the chip into quadrants and splitting the light into four
> paths
>> that each pass through a different emission filter before hitting the
> CCD.
>> 
>> Just clarify, I have never personally used one of these in a
> research
>> project, but I have worked with one in developing software support
> for the
>> resulting images. I do not work for Photometrics.
>> 
>> Chris Tully
>> Microscopy and Image Analysis Expert
>> [log in to unmask]
>> 240-475-9753 (c)
>> 
>> [image: View my profile on
> LinkedIn]<http://www.linkedin.com/in/christully/>
>> 
>> 
>> On Tue, Mar 5, 2013 at 2:09 PM, Jean-Pierre CLAMME < 
>> [log in to unmask]> wrote:
>> 
>>> *****
>>> To join, leave or search the confocal microscopy listserv, go to:
>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>> *****
>>> 
>>> Hi Andreas,
>>> 
>>> The issue is not grouping channels in the sequential box and
> choosing line
>>> and frame. The issue is changing filter sets quickly. To take the 3
> images
>>> DD, DA, AA the filter/light pathway has to be changed between
> images. DD
>>> and DA  or DD and AA can be taken in the same configuration, but AA
> and DA
>>> can't. So the only way I can see is to use the Virtual channels and
> that
>>> is too slow line by Line.
>>> 
>>> Best
>>> 
>>> JP
>>> 
>>> 
>>> 
>>> 
>>> 
>>> Confocal Microscopy List <[log in to unmask]> wrote
> on
>>> 03/05/2013 07:01:06 AM:
>>> 
>>>> Andreas Bruckbauer <[log in to unmask]> Sent by: Confocal Microscopy 
>>>> List
> <[log in to unmask]>
>>>> 
>>>> 03/05/2013 07:05 AM
>>>> 
>>>> Please respond to
>>>> Confocal Microscopy List <[log in to unmask]>
>>>> 
>>>> To
>>>> 
>>>> [log in to unmask]
>>>> 
>>>> cc
>>>> 
>>>> Subject
>>>> 
>>>> Re: Ratiometric FRET on Fluoview
>>>> 
>>>> *****
>>>> To join, leave or search the confocal microscopy listserv, go to:
>>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>>> *****
>>> 
>>>> I have not used it for FRET, but we have a sequential tick box at 
>>>> the bottom of the window with the channel parameters. When ticked
> we
>>>> can select "frame by frame" or "line by line" aquisition and how
> the
>>>> channels will be grouped together.
>>> 
>>>> best wishes
>>> 
>>>> Andreas
>>> 
>>>> 
>>>> -----Original Message-----
>>>> From: Jean-Pierre CLAMME <[log in to unmask]>
>>>> To: CONFOCALMICROSCOPY <[log in to unmask]>
>>>> Sent: Mon, 4 Mar 2013 22:44
>>>> Subject: Ratiometric FRET on Fluoview
>>> 
>>>> 
>>>> *****
>>>> To join, leave or search the confocal microscopy listserv, go to:
>>>> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
>>>> *****
>>> 
>>>> Hello,
>>> 
>>>> I saw a paper about Ratiometric FRET (Roszik, Cytometer 2009)
> mentioning
>>>> line by line acquisition of the 3 images (IDA, IDD, IAA). The
> authors
>>>> mentioned the use of a LSM 510.
>>> 
>>>> I don't know the LSM510, but I have a fluoview 1000 and the only
> way I
>>> can
>>>> see how to do that is using the "virual Channels" . However that
> would
>>> be
>>>> image by image and not line by line.
>>> 
>>>> Does anyone has done ratiometric FRET on the fluoview and what
> method
>>> did
>>>> you use ?
>>> 
>>>> Thank you and Best regards,
>>> 
>>>> JP
>>> 
>>>> 
>>>> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> -
>>>> Jean-Pierre CLAMME, PhD
>>>> Chief Scientist
>>>> Nitto Denko Technical
>>>> 501 Via Del Monte
>>>> Oceanside, CA 92058
>>>> E-mail: [log in to unmask]
>>>> Phone: 1-760-696-9428
>>> 
> 
> ______________________________________________________________________
> The information in this email is confidential and intended solely for the addressee.
> You must not disclose, forward, print or use it without the permission of the sender.
> ______________________________________________________________________