Test Lightroom/Photoshop Real world usage - SSD's

[JasonSTL739]

I have an unusual opprotunity to compare a large number of different disk subsystems. I'm curious how the community would think best to go about a non-biased test for Lightroom/Photoshop usage.

The goals are:
-Show SSD's are worth it
-Show which of the X25-E, X25-M, and OCZ Vertex's are the best bang for the buck.
-How how RAID0 increases the above.

I plan to compare to a few standard disks, including:
-Two SAS disks on RAID0
-Eight SAS disks on RAID0
-Two Raptors RAID 0
-Two 1.5TB Seagates

I have ton of SSD's available for this little test. I can create a 12-disk X25-E array, for example, for fun. One key test I want to show is four of each against each other in benches.

Need suggestions. I can play around with various import/export/actions, but I realize my workflow might be very diferent than other people's.

Some initial thoughts:
-import of images into lightroom
-Creation of previews
-Some "actions" moving around inside lightroom
-Export from lightroom
-Editing an image (creating a .tif) to open an image in photoshop
-A few actions inside photoshop
-Saving the file after a set series of actions.

Thoughts?

[ed rader]

I have an unusual opprotunity to compare a large number of different disk subsystems. I'm curious how the community would think best to go about a non-biased test for Lightroom/Photoshop usage.

The goals are:
-Show SSD's are worth it
-Show which of the X25-E, X25-M, and OCZ Vertex's are the best bang for the buck.
-How how RAID0 increases the above.

I plan to compare to a few standard disks, including:
-Two SAS disks on RAID0
-Eight SAS disks on RAID0
-Two Raptors RAID 0
-Two 1.5TB Seagates

I have ton of SSD's available for this little test. I can create a 12-disk X25-E array, for example, for fun. One key test I want to show is four of each against each other in benches.

Need suggestions. I can play around with various import/export/actions, but I realize my workflow might be very diferent than other people's.

Some initial thoughts:
-import of images into lightroom
-Creation of previews
-Some "actions" moving around inside lightroom
-Export from lightroom
-Editing an image (creating a .tif) to open an image in photoshop
-A few actions inside photoshop
-Saving the file after a set series of actions.

Thoughts?


dude....what language are you speaking :D?

aren't SSDs like the clap or sumthin' :D?


ed rader

[JasonSTL739]

dude....what language are you speaking :D?

aren't SSDs like the clap or sumthin' :D?


ed rader

Heh. :cool::p

[Grentz]

Honestly the problem is the times you are dealing with are going to be tiny. I run with a single Raptor 150gb drive and work with my photos on a standard 300gb SATA drive and lightroom loads up quick and renders previews almost instantly as well as applying things even quicker. If anything more stress is put on the CPU than the disks with lightroom.

[JasonSTL739]

Honestly the problem is the times you are dealing with are going to be tiny. I run with a single Raptor 150gb drive and work with my photos on a standard 300gb SATA drive and lightroom loads up quick and renders previews almost instantly as well as applying things even quicker. If anything more stress is put on the CPU than the disks with lightroom.

Will be fun to see - I do not think that is the case based on my experience with working with Photoshop and a set of 30GB Vertex's on this Win7 machine.

One of the more interesting tests will be a Vista x64 virtual machine running entirely on a RAM disk. The OS, Lightroom, Photoshop, swaps, image files - all of it. We'll see how it deals with that.

[Wilt]

Too many variables! The fastest would be if Lightroom cache was on SSD and so was the library. The next fastest would be if Lightroom cache was on SSD and the library was on RAID...the slow part would be loading from library into cache, but once cache was loaded the edits would be fast. The next fastest would be if cache was on local drive, and the library was on RAID, as the transfer would be slow and so would edits. The next slower would be cache on local drive, and data on RAID. The slowest would be all on the same harddrive.

[Grentz]

Will be fun to see - I do not think that is the case based on my experience with working with Photoshop and a set of 30GB Vertex's on this Win7 machine.

One of the more interesting tests will be a Vista x64 virtual machine running entirely on a RAM disk. The OS, Lightroom, Photoshop, swaps, image files - all of it. We'll see how it deals with that.

That is photoshop though, not lightroom ;)

Photoshop most definitely gets huge gains when working with very large files through fast scratch disks and working disks. Lightroom on the other hand seems pretty snappy with most of its edits and previews even on a simple machine by todays standards (it is quick on my laptop with a C2D T7300 @ 2ghz)

[JasonSTL739]

Too many variables! The fastest would be if Lightroom cache was on SSD and so was the library. The next fastest would be if Lightroom cache was on SSD and the library was on RAID...the slow part would be loading from library into cache, but once cache was loaded the edits would be fast. The next fastest would be if cache was on local drive, and the library was on RAID, as the transfer would be slow and so would edits. The next slower would be cache on local drive, and data on RAID. The slowest would be all on the same harddrive.

Agreed - mainly listing the options.

What I'm after mainly is the i7 system with the Areca 1680; with various flavors of disk hanging off of that.

The R905 is mainly to test the whole thing "in a VM"

I've found (as have others) that the cache and library doesn't effect things as much as you'd expect. I'm running that today on the 4 30GB Vertex's and the image file location is also very key.

That is photoshop though, not lightroom ;)

Photoshop most definitely gets huge gains when working with very large files through fast scratch disks and working disks. Lightroom on the other hand seems pretty snappy with most of its edits and previews even on a simple machine by todays standards (it is quick on my laptop with a C2D T7300 @ 2ghz)

I'm not sure all would agree - but then again speed is subjective I suppose.

[Grentz]

I'm not sure all would agree - but then again speed is subjective I suppose.

I guess, and both my machines are pretty fast that I do editing on. Lightroom edits just are not that intensive for the most part.

[hollis_f]

I have an unusual opprotunity to compare a large number of different disk subsystems. I'm curious how the community would think best to go about a non-biased test for Lightroom/Photoshop usage.

The goals are:
-Show SSD's are worth it

So this is a proper, scientific, unbiased test? Surely the goal should be to see if SSDs are worth it.

[JasonSTL739]

So this is a proper, scientific, unbiased test? Surely the goal should be to see if SSDs are worth it.

Trying for it to be, at least for the standard benches that I'm working through right now (same system, same benches, same config/aligns) Ignoring capacity SSD already just kill everything out there for the $ (Vertex's especially) I came from a bunch of SAS disks to the Vertex's in my main workstation and there is no contest unless you are working with video. It just feels different - everything is fast(er) by a large margin.

We'll see. Another forum member on a different forum lead me to this for photoshop:
http://www.retouchartists.com/pages/speedtest.html

Which should help test the overall system with different configs.

[JasonSTL739]

Too many variables! The fastest would be if Lightroom cache was on SSD and so was the library. The next fastest would be if Lightroom cache was on SSD and the library was on RAID...the slow part would be loading from library into cache, but once cache was loaded the edits would be fast. The next fastest would be if cache was on local drive, and the library was on RAID, as the transfer would be slow and so would edits. The next slower would be cache on local drive, and data on RAID. The slowest would be all on the same harddrive.

I would have to agree with you - too much time to get this done as it relates to photoshop/lightroom performance, as there is quite a bit of tuning that can be done with each flavor of disk.

Here is some data.

This is entirely on the ICH10R, pretty much any drive I could hang off of it. 1X, 2X, 3X, etc. Without question, it is limited to around 600MB/sec

-Intel Matric Cache was always off. It DEFINITELY has a huge effect on writes especially, but for the purposes of stable testing I didn't mess with it.

-Always 128k strip on RAID.
-Aligned to 128 if it was an SSD.

I killed the ICH10R on one of the two i7 board I have before I realized the northbridge/southbridge pipe assembly needed additional cooling. The numbers in this test are a second run after I actively cooled the chipset. I was getting some STRANGE numbers!

Spreadsheet only:
http://www.sedura.com/linking/benching/ICH10Rresults.xlsx (http://www.sedura.com/linking/benching/ICH10Rresults.xlsx)

Zip file with spreadsheet and all screenshots (8MB)
http://www.sedura.com/linking/benching/ICH10Rresults.zip (http://www.sedura.com/linking/benching/ICH10Rresults.zip)

[JasonSTL739]

For fun...

These kind of benches are pretty worthless for this type of disk subsystem, but here it goes. I'll be playing extensively with IOMeter, SQLIO, and Benchmark Factory with the 1680 and the X25s in weeks to come to show and tell.

All X-25-E's (12) on Areca 1680.

http://sedura.smugmug.com/photos/510140477_MjNMc-M.jpg

http://sedura.smugmug.com/photos/510140480_AVtqa-M.jpg


5 X25-E's on the ICH10R:
http://sedura.smugmug.com/photos/510140243_9KJ2U-M.jpg

http://sedura.smugmug.com/photos/510140253_rBfZw-M.jpg

[JasonSTL739]

For those that want to know about photoshop, here is a better test than what I have hardware for:
http://download.intel.com/design/flash/nand/extreme/Photoshop_CS4_Performance_Comparison.pdf (http://download.intel.com/design/flash/nand/extreme/Photoshop_CS4_Performance_Comparison.pdf)

Too many variables on my side - I decided not to waste the time.

Photoshop and Lightroom's differences with SSD are night and day in use - I will not be going back, EVER. I'm selling off all the SATA disks except for the 1.5TB drives pronto.

Especially in a laptop! Wow what a difference in actual use.

[JasonSTL739]

Here is another interesting exercise. Vertex firmware revisions - all four of them - on a 120GBer. Firmware *matters* on SSD even moreso than HD's.

Look at the difference at the firmware most of the reviews out there are based on versus the newest!!!!

Original 0112:
http://sedura.smugmug.com/photos/509676146_AhCKG-M.jpg

1199:
http://sedura.smugmug.com/photos/509676159_YBz73-M.jpg

1275:
http://sedura.smugmug.com/photos/509676173_xpQco-M.jpg

1370:
http://sedura.smugmug.com/photos/509676187_9deur-M.jpg

[vinceman]

subscribed. thanks.

Jason, you are on to something here.

SSDs are becoming more and more mainstream. For a first time would be buyer like me, I would wanna know if it will improve my Lightroom experience.

What kind of setup is optimal.. ie catalog on hdd, cache on ssd, or both on single ssd. Would there be significant difference if I make a striped array?

If we can quantify the improvement, potential users will be able to weigh in if the price is worth that improvement.

[JasonSTL739]

subscribed. thanks.

Jason, you are on to something here.

SSDs are becoming more and more mainstream. For a first time would be buyer like me, I would wanna know if it will improve my Lightroom experience.

What kind of setup is optimal.. ie catalog on hdd, cache on ssd, or both on single ssd. Would there be significant difference if I make a striped array?

If we can quantify the improvement, potential users will be able to weigh in if the price is worth that improvement.

Most of the "feel" for me comes from the OS being installed on the SSD's. However, I now run the catalog, cache, and previews there. Makes too much sense. I run the actual images from a secondary drive set (RAID10, four 1.5TB Seagates) usually. Expect for import/export/save operations, there isn't much of a benefit to the images on the SSD's compared to the bulk array with a good RAID card.

[Kuma]

Interesting topic but doesnt' Lightroom and Photoshop perform well enough for you already? I dont even have a stellar PC and both these packages (especially Lightroom) are blazing fast already. I'm using Lightroom more than Photoshop these days and I can't recall a single instance when I thought that I needed more speed. :rolleyes:

[JasonSTL739]

Interesting topic but doesnt' Lightroom and Photoshop perform well enough for you already? I dont even have a stellar PC and both these packages (especially Lightroom) are blazing fast already. I'm using Lightroom more than Photoshop these days and I can't recall a single instance when I thought that I needed more speed. :rolleyes:

How many images - and what size - are you working with?

Speed is a relative thing, for us it is never fast enough. We're working with 300-400MB files in photoshop, and 25MB or so RAW files, but also with Illustrator, Lightroom, Outlook, IE, burning a disk, etc all running at once on a system.

There is a "crispness" with SSD you have to experience to see the value I suspect. I will never go back to an OS disk that is non-SSD.

[Bobster]

what happens to the performance of the system after a week, 2 weeks a month?

[JasonSTL739]

what happens to the performance of the system after a week, 2 weeks a month?

I've been running Vertex's in a workstation and a laptop, and also a set of X25's in a second workstation, and they are all still nice and fast.

The "drop" in performance for real-world usage for these two model of SSD's are highly overrated. A JMitron-based SSD with no cache will see it, but these will not.

Also - there is a TRIM tool for the Vertex's already that pretty much addresses any concern. TRIM is also to be in Windows 7, and many think it is already in the RC.

[tuan209]

There is a new firmware for the intels that fixes the performance degradation.