Benchmark Photoshop Speed Test, V3 for really fast computers!

[HankScorpio]

If you want a new standard test to use that's a bit more taxing for newer machines then the Driver Heaven Photoshop Benchmark V3 (http://www.driverheaven.net/photoshop.php) is probably one of the most comprehensive around at the moment. Mac and PC compatible from PS7 up to CS4.

The image is included and is 109MB uncompressed jpeg so only 12MB on disc.

[HankScorpio]

I've just run it and get 236.8 seconds with:

Intel Core i7 920 (stock speed)
12GB DDR3 RAM

Photoshop CS4 64bit
Vista Home Premium 64bit

[Bwaze]

Driver Heaven Photoshop Benchmark V3 (http://www.driverheaven.net/photoshop.php)

Windows 7 64bit:
Intel i920 @2,66 GHz, Asus P6T, 6 GB RAM, GeForce 8800 GTS: 241s

Maybe start a new thread for this test?

[Moppie]

It seems the old Benchmark test was realising the end of its useful life, with times getting close to 10 seconds.

It appears curent generation i7 rigs are so power ful it is hard to get a meaningful comparison from it.


So, here is a new test, which use a larger file, and does more things, and it will take even a dual i7 rig a little more than 10 seconds to complete.

[Pete]

Mac Book Pro 2.6Ghz, 4GB RAM

408.6s

Oh well, it's just a laptop.

[Moppie]

My Q6600 with 4GB of ram on an intel Dragontail m/b

323sec's.

Never used more than 27% of the CPU (all 4 cores combined), or 2.03GB of ram though.

The other speed test would hit 100% of each, seperatly, and together depending on what it was doing.

[John_TX]

This test is bogus, it hardly uses more than one processor.

[HankScorpio]

This test is bogus, it hardly uses more than one processor.
Why is that bogus? Retouching an image in PS barely gets my CPU excited enough to register 10% on one core.

All synthetic tests are kind of pointless but they are good for measuring your e-manhood against others.

[basroil]

This test is bogus, it hardly uses more than one processor.

Well, so do most filters in PS. It may be better if the test ran two or three (or more) runs at once, but that's the nature of "unbiased" testing.


All synthetic tests are kind of pointless but they are good for measuring your e-manhood against others.

So true... my e-manhood is 2.534% larger than yours ;)

[tommykjensen]

Finally tested my new machine. The test failed on my old so have nothing to compare with.

- Gigabyte GA-X58-UD4 motherboard
- Core i7-920 4x2.66GHz 4,8GT/sec
- 12 GB DDR3 TriChannel(1333MHz)
- 2 x Seagate 1500GB S-ATAII 3Gb/s 32mb
- 2 x WD Raptor 150GB 10000rpm (one for Vista and one for ps scratch disk)
- Pioneer Blu-Ray Combo BDC-202BK
- Nvidia GTX295 1792MB PCI-E
- SilverStone KL01B

Windows Vista Ultimate 64 bit

253,3 seconds.


My new machine is a breeze to work on. Everything in my workflow that on my old takes>5 seconds is done in <1 second now. Something simple as starting PS CS4 now takes <2 seconds on the old it takes about 27 seconds. Opening an image in camera raw usually took several seconds its now done instantly.

[Bobster]

3Ghz Dual Core - 8GB RAM, Win7 64

usual shizzle running - CaptureOne in the background processing some images from yesterdays wedding, Winamp blaring, Firefox with this page and a few extras running..

319.9 Seconds

[lifethroughalens]

Inspiron 9400 - 2.16GHz Core 2 Duo- 2GB RAM - 256 ATi - W7 RC

514.1 seconds - Quite respectable for a 4 year old consumer lappy :)

[Yogesh Sarkar]

It took 356 seconds on my C2D E4500 + 2x1GB 667mhz ram and 160GB SATA 2 HDD. Not bad for a rig that I put together for less than US$ 500, one and a half year back (computer parts are costlier in India).

[Bobd314]

Weird.. I feel like my computer is kind of old, and it only took me 257.3 seconds.

2 GB Ram
Core2Duo E6850 @ 3.0 GHz
Windows 7 64

[Moppie]

Weird.. I feel like my computer is kind of old, and it only took me 257.3 seconds.

2 GB Ram
Core2Duo E6850 @ 3.0 GHz
Windows 7 64


Yip, look at the results compared to the other test, it is clearly not a good one.

I got supsecious when it never used all of my systems resources, while the other test pushed everything to 100%.

[F4 Cyborg]

Just for Grins.
Everything normal running (why not it's how I do it any other time)

NoteBook System
Asus G50VT
Intel® Core™2 Duo Processor T9550@2.66ghz 2.67ghz, 1066MHz, 6.MB On-Die L2 Cache;
Mobile Intel® PM45 Express Chipset +ICH9M
400G(200Gx2),7200rpm;
Vista Ult 64 bit
X's 2 (one for backup)
Already has the boost prog to just click and ride.
and of course external HD's to back up files.

Maybe I oughta go thru CS3 and set things to run faster.

And since I use CS3 it is 32 bit, but LR@64bit you gota love it supersonic speed.

With out Boost - 292.15
Boosted to 3.1 - 245.5

I went to Notebooks well why not? I'm always somewhere else.

(the only thing I have added to the lineup is a "Cool Master" USB powered cooling tray, to help keep the chips Cool)
Only large fault I ever shrugged NoteBook's for was ability to stay cool.
If any of you have a better Idea (portable) I'd be happy to hear it.

[twofruitz]

Finally tested my new machine. The test failed on my old so have nothing to compare with.

- Gigabyte GA-X58-UD4 motherboard
- Core i7-920 4x2.66GHz 4,8GT/sec
- 12 GB DDR3 TriChannel(1333MHz)
- 2 x Seagate 1500GB S-ATAII 3Gb/s 32mb
- 2 x WD Raptor 150GB 10000rpm (one for Vista and one for ps scratch disk)
- Pioneer Blu-Ray Combo BDC-202BK
- Nvidia GTX295 1792MB PCI-E
- SilverStone KL01B

Windows Vista Ultimate 64 bit

253,3 seconds.




Mine shouldn't be anywhere near as fast as yours, wondering if its due to CS3/CS4 or Vista/7.

- Gigabyte GA-X58-UD3R motherboard
- Core i7-920 2.66ghz at 2.8ghz
- 6GB DDR3 Corsair TriChannel(1600MHz)
- 640gb Western Digital Caviar Black 32mb Cache 7200RPM HDD.
- ATI 4870 1gb Graphics

- Windows 7 RC
- Photoshop CS3 32bit

265 seconds.

[GCGuy]

Toshiba laptop
Intel T6400 2 Ghz
4GB DDR2
CS4 64bit

378.6 seconds

CS4 32 bit

414.3 seconds

[paavor]

HP "el Cheapo" laptop, PS CS4 / Vista 32bit


AMD Sempron SI-40, 2000 MHz (Single core)
nVIDIA GeForce 8200M
2816 MB DDR2-667 (actually 4 GB installed..)
NVIDIA GeForce 8200M G (256 MB)


1257,1 sec :mrgreen: (I just had to try..)

Maybe someday I'll benchmark my workstation too, but this was much more fun. :p

[Moppie]

I'm going to unsticky this, as the test seems to be a rather poor indication of true computer performance.

I will gladly sticky a new thread if someone can come up with a better test.

[lifethroughalens]

Just upgraded my PC:

W7 64bit Evaluation Copy: 7100

i7 920 OC'd to 3.4Ghz on stock cooler
6GB Patriot DDR3 1600Mhz RAM
Asus P6T Deluxe V2 Motherboard
1TB 7.2K 32MB WD Caviar Black HDD
Corsair 650W TX PSU
1x Nvidia GTX260 GPU
Coolermaster Sniper Case
CS3 Extended (I think it would be significantly quicker in CS4 sine this would utilise the GPU better)

234 seconds flat. (Sure beats the hell out of my old Core 2 Duo system!)

[Mike]

I'm going to unsticky this, as the test seems to be a rather poor indication of true computer performance.

I will gladly sticky a new thread if someone can come up with a better test.

I have to agree. I ran it on CS2 on a Win Vista 64 bit, Intel Quad core Q9550, 8GB ram system and got 333.1 seconds...which doesn't seem all that quick compared to other systems...(or is it me?!!) :confused:

[gjman]

I have to agree. I ran it on CS2 on a Win Vista 64 bit, Intel Quad core Q9550, 8GB ram system and got 333.1 seconds...which doesn't seem all that quick compared to other systems...(or is it me?!!) :confused:

I get 469 sec for CS2 on Q6600/8GB running Vista Ultimate x64 SP1 :)

[MaxxuM]

I have to agree. I ran it on CS2 on a Win Vista 64 bit, Intel Quad core Q9550, 8GB ram system and got 333.1 seconds...which doesn't seem all that quick compared to other systems...(or is it me?!!) :confused:

Yea, not a great benchmark. My PC is the same as yours, faster HDD and video probably, but same CPU and memory. I have CS3 on my PC and I got a 269.9 with no tweaking or changes to PS. About a minute difference between us just because I have CS3? Doesn't make sense.

[lens pirate]

202.5

Core i7 oc'd to 3.8
12 gigs ram
5 gigs of ram used as ram drive with photoshop scratch file assigned to the ram drive
Nvidia SLI 2ea 280 video cards
Adobe CS4 64 bit
Vista 64 bit

The ram drive seemed to really help

[RockSlut]

297s

MB: ASUS K6T
CPU: Intel i7 920 (stock)
RAM: 12gb Kingston 1333
HDD 1: 60GB Vertex series OCZ SSD
Raid 0: 2 x WD Caviar Green 1.5TB 7200rpm
Wireless PCI: ASUStekWL-138G v2
Video Card: Galaxy 9800GTX
PS CS3
Windows 7 RC 7100

Was expecting a better score, but the PC is lightning fast for my needs regardless.

[tester3000]

161.4s

System is in my sig.

[beeng]

Core 2 Duo P8700.
3Gb RAM
ATI Radeon 3670

293.2 Seconds
on Photoshop CS4 x64, Windows 7 x64

[lens pirate]

161.4s

System is in my sig.

Love to hear about your system settings and cooler thaty lets you run 4.2 on that 920. I can get 4.0 but I have heat issues and eventually running hard-core CPU burn in apps I get blue screens. Never when using justing the PC. On the other hand at 3.8 I can control the heat and never fail any tests. Any congrates on a killer system!

[tester3000]

I'm using a Thermalright Ultra 120 eXtreme, with 2 1600rpm fans in a push/pull config. Air goes out the top of my Antec 900 (200mm fan on top).

I'm running 200x21(turbo boost) @ 1.3v /w Hyperthreading on. If it turn HT off i can lower my volts to 1.25v. QPI is at 1.34v and ram at 1.66v. Temps are a touch on the high side, hitting 70oC with MAX load, they usually stay around 40-50oC. I guess i just got a gem of a cpu :D And thank you :)

[lens pirate]

I'm using a Thermalright Ultra 120 eXtreme, with 2 1600rpm fans in a push/pull config. Air goes out the top of my Antec 900 (200mm fan on top).

I'm running 200x21(turbo boost) @ 1.3v /w Hyperthreading on. If it turn HT off i can lower my volts to 1.25v. QPI is at 1.34v and ram at 1.66v. Temps are a touch on the high side, hitting 70oC with MAX load, they usually stay around 40-50oC. I guess i just got a gem of a cpu :D And thank you :)

I wish.... if I ran cpu volts 1.3 My rig would melty down. My Cool master V8 just can't deal with it. I KNOW I am limited by heat. I should upgrade my cooler. Currently I run 1.26 on my CPU, 1.35 QPI and 1.65 on the ram. My temps under full load never exceed 70c. I am talking linpac not that weak prime 95 test either. IF I could run 1.3 volts on the CPU I am sure I could run 4+ ghz safely.

[beeng]

Needs more liquid nitrogen.

[basroil]

Needs more liquid nitrogen.

Not sure you need that, but some polydiamond dust in the thermal grease+ phase change cooling would keep it cold enough without risking thermal fatigue ;)

[beeng]

Not sure you need that
Is there ever a need for bacon? No, but sometimes you just have to go all out. :p

[ZGMF-X20A]

Stock: 255.4
OC: 171.3
OC yields 32.9% improvement over the old timing
System: i7 920 OC to 4.0GHz, 12GB RAM

[lens pirate]

Stock: 255.4
OC: 171.3
OC yields 32.9% improvement over the old timing
System: i7 920 OC to 4.0GHz, 12GB RAM


GREAT DATA!!

[ZGMF-X20A]

GREAT DATA!!
Thanks

I wish.... if I ran cpu volts 1.3 My rig would melty down. My Cool master V8 just can't deal with it. I KNOW I am limited by heat. I should upgrade my cooler. Currently I run 1.26 on my CPU, 1.35 QPI and 1.65 on the ram. My temps under full load never exceed 70c. I am talking linpac not that weak prime 95 test either. IF I could run 1.3 volts on the CPU I am sure I could run 4+ ghz safely.

I actually uses the Cooler Master V8 and able to run my CPU @ 1.35, QPI @ 1.35 & RAM @ 1.66. I used the CPU stress test (prime95 (http://files.extremeoverclocking.com/file.php?f=205)) for about an hour without any problem. In conjunction, I'm also running GPU stress test (Real-Time High Dynamic Range Image-Based Lighting (http://www.daionet.gr.jp/~masa/rthdribl/)) too.

I left the computer running while watching TV, grab snack then come check the computer again. I don't have the exact ambient temperature, but we have the heater on in the room at that time.

Under stress, my CPU has a max of 91C and GPU has a max of 75C. I read that i7 should be able to handle low 91C with ease. What mobo are you using? I use the P6T-SE.

[bsmotril]

Old School core2Quad Q9600 @2.66GHz with 8gig Ram / 6 gig avail to CS4-64bit gave me 275.4 seconds.

[lens pirate]

When I first ran this test under Vista 64 I got 202.5 seconds.

Now running Windows 7 all machine specs exactly the same, with the only change being Windows 7 I scored 180.7

That is a 10 percent improvement in speed. bw!

[lens pirate]

When I first ran this test under Vista 64 I got 202.5 seconds.

Now running Windows 7 all machine specs exactly the same, with the only change being Windows 7 I scored 180.7

That is a 10 percent improvement in speed. bw!

I am surprised nobody commented on this, to me it is very interesting and unprecedented. Imagine a O/S from of all people Microsoft that runs FASTER than the previous ones while being more stable, feature rich, and very easy to use.

Nobody else is amazed by this turn of events?

[Moppie]

These tests are not 100% accurate, or the same every time.

I ran it again 12 months after first running it, on the same Vista install, and it was about 10% faster.

[lens pirate]

These tests are not 100% accurate, or the same every time.

I ran it again 12 months after first running it, on the same Vista install, and it was about 10% faster.

No other changes? Same video drivers? That is interesting as I have run this test over and over and find the results very consistant. PLus or minus a few tenths of a second test to test.

[ZGMF-X20A]

I am surprised nobody commented on this, to me it is very interesting and unprecedented. Imagine a O/S from of all people Microsoft that runs FASTER than the previous ones while being more stable, feature rich, and very easy to use.

Nobody else is amazed by this turn of events?

I think most review has indicated this. Some subjective even thinks that Win7 is faster than XP on some tasks although measured test indicates that win7 is slightly slower than XP.

[Moppie]

No other changes? Same video drivers? That is interesting as I have run this test over and over and find the results very consistant. PLus or minus a few tenths of a second test to test.

No changes at all. :cool:

[basroil]

No changes at all. :cool:

a few "small" updates though ;)

[ZGMF-X20A]

Moppie, I'd say the 10% drop in a year can be attributed to a lot of things. Here's a few that I can think of:
1. Scratch disk is on a disk with fragments
2. Registry bugged down
3. Other software installed on the system

12 months is a long time. Slow down is usually expected on a machine that's used regularly.

[Moppie]

It wasn't slow down. The time improved.

The system has had no hardware changes, and only the constant stream of regular MS updates.

There are a lot of variables at work, maybe the scratch disk used a different part of the disc that it could read and write from faster?
Maybe it was cooler and the CPU ran faster?
Who knows.

All I know is the same hardware with the same install was faster 12 months later.
Not a lot faster, but enough to show that there can be variation in the test results.

[lens pirate]

It wasn't slow down. The time improved.

The system has had no hardware changes, and only the constant stream of regular MS updates.

There are a lot of variables at work, maybe the scratch disk used a different part of the disc that it could read and write from faster?
Maybe it was cooler and the CPU ran faster?
Who knows.

All I know is the same hardware with the same install was faster 12 months later.
Not a lot faster, but enough to show that there can be variation in the test results.

You might be on to something with the scratch disk. My photoshop scratch disk is on a compressed 4 gig ram disk. That is used only for photoshop scratch. Perhaps that explains my results being repeatable.

[bohdank]

Win 7 64 bit

32 bit PS 231.8
64 bit PS 215.7

E8400 dual core overclocked to 3.6
6 gigs memory
9800GT 500megs

[dashotgun]

210.2
win 7 64 photoshop cs4
i7 860 2.8 oc to 3.4
8 gb ram ddr2 1600

[lifethroughalens]

W7 64bit Home

i7 920 OC'd to 3.4Ghz on stock cooler
6GB Patriot DDR3 1600Mhz RAM
Asus P6T Deluxe V2 Motherboard
1TB 7.2K 32MB WD Caviar Black HDD
Corsair 650W TX PSU
1x Nvidia GTX260 GPU
CS4 Extended

206.4 (Improved from CS3 & W7 Beta @ 234 secs)

[lens pirate]

You know people poopooed this test...but it seems to scale pretty well accross CPU speeds and types.

Why is this test not useful again as a benchmark for overall Photoshop performance on a given machine?

Is there a better tool for judgeing a specific computer config against photoshop?

[Moppie]

You know people poopooed this test...but it seems to scale pretty well accross CPU speeds and types.

Why is this test not useful again as a benchmark for overall Photoshop performance on a given machine?

Is there a better tool for judgeing a specific computer config against photoshop?


It doesn't use the full 100% of system resources, and it doesn't make gooduse of multiple cores as it relies to much on old filters and process.
Hence a couple of dual core machines have out performed better spec quad cores.

This test: http://photography-on-the.net/forum/showthread.php?t=170063
Make much better use of all resources and includes filters and process's that will use 4 or more cores.
Unfortunately the times are getting very, very low and its getting harder to get a meaningful comparison.

[lens pirate]

It doesn't use the full 100% of system resources, and it doesn't make gooduse of multiple cores as it relies to much on old filters and process.
Hence a couple of dual core machines have out performed better spec quad cores.

This test: http://photography-on-the.net/forum/showthread.php?t=170063
Make much better use of all resources and includes filters and process's that will use 4 or more cores.
Unfortunately the times are getting very, very low and its getting harder to get a meaningful comparison.

I hear what you are saying.

Still a reasonable person can READ PC specs and from that make a reasonable set of assumptions about performance.

If a test then confirms those assumptions along with other checks it seems to me to be a reasonable measure. Reading through this thread this test seems to track obvious assumptions one could make about CPU speed and over all system speed.

We have systems here in this thread ranging from 160 seconds on up through hours. I bet if you plotted system specs and results on a graph there would be no real suprises.

I bet the results would be as useful as that other test for comparing your machine against others and maybe using that information to drive a purchase.

[tt350z]

Laptop w/ all my normal stuff running
Nvidia Geforce 9800M GT
Core 2 Duo
E8600 @ 3.33GHZ
3.25GB ram DDR2 PC5300
XP Pro
CS4

254.3

[gravy graffix]

ugh 460 sec...
D940 3.2 ghz
2 gig of slow ram, my good ram commited suicide.
w7 64 bit
cs4
8800GTS SLI (just one tho)

[MikeC710]

ugh 460 sec...
<snip>

my good ram commited suicide

<snip>

Ha! My first chuckle of the day...thanks!

[MikeC710]

209.5 seconds on a 3 day old -
Studio XPS 8100, Intel Core i7-860 processor(8MB Cache, 2.80GHz)
6GB DDR3 SDRAM at 1333MHz-2X2GB/2x1GB
AMD Radeon 5450 (1GB)
500GB Serial ATA 2 Hard Drive 7200 RPM
Windows 7 Home Premium, 64bit
- running CS4.

I'll be intersted to see how a few months of scrambling the silicon on this machine will effect the results.

[akiwi]

Ha, looks like I'm the only person on this forum who opted for an i5 processor!
I saw that the photoshop performance was almost exactly the same as the i7, but it cost less. So I'm happy with my result.
Packard Bell iExtreme M5740
Processor: Intel cor i5 CPU 750 @ 2.67 GHz
Ram: 6.0 GB
Win 7 64
CS4 64

259.1 seconds

[Fuzzmuffin]

331.9 seconds

iMac 21.5" Core 2 Duo
3.06GHz 8GB RAM
CS3

[Tony-S]

Ha, looks like I'm the only person on this forum who opted for an i5 processor! I saw that the photoshop performance was almost exactly the same as the i7, but it cost less. So I'm happy with my result.

That's because Photoshop can't do hyperthreading. You want fast, you should try Aperture or Pixelmator. ;)

[StudioBin]

For the people who OC their CPU. How stable is it? Do you shut off your computer when you are not using it? I normally do not turn my computer off. I just leave it on 24/7 365.

[lifethroughalens]

For the people who OC their CPU. How stable is it? Do you shut off your computer when you are not using it? I normally do not turn my computer off. I just leave it on 24/7 365.

Had my i920 OC'd to 3.6 for months now - on like yours, almost always.

I did have an issue with RAM voltages to begin with - my RAM said it couldn't take any more than 1.65V but my bios would only let me OC in multiples of 0.02 - So I tried 1.66V and had two occasions over a couple of months when the OS failed to load, so I dropped down to 1.64V and now it's been absolutely great for months.

Interestingly using the stock cooler, I noticed next to no increase in idle or load temps before and after OCing. I use a cooler master case set on the lowest fan settings.

So just make sure you don't over power the RAM, keep half an eye on temps, and you'll be fine.

[osmaneralp]

178.2 seconds with the following setup:

Windows 7 64 bit
i7 920 @ 3.8 GHz
6 GB RAM
Intel SSD as primary drive and PS swap drive

@StudioBin: my overclocked setup is stable running Prime95 for 24+ hours; however, I put my system to sleep when not using it to save energy.

FWIW, I get 240.3 seconds when running my i7 at the stock 2.66 GHz.

--Osman

[lifethroughalens]

178.2 seconds with the following setup:

Windows 7 64 bit
i7 920 @ 3.8 GHz
6 GB RAM
Intel SSD as primary drive and PS swap drive

@StudioBin: my overclocked setup is stable running Prime95 for 24+ hours; however, I put my system to sleep when not using it to save energy.

FWIW, I get 240.3 seconds when running my i7 at the stock 2.66 GHz.

--Osman

Hey Osman, what SSD are you using with your set-up? Looks like your getting a 10% advantage over my 3.6 OC with 7.5K HD's. I'll be making a move over to SSD's for the Boot and swap drives soon. Cheers, Rod

[StudioBin]

osmaneralp.. thx for the info. I am on the fence on a custom PC or going iMac for the small form factor. On one hand I will have better hardware specs., on the other the form factor of the iMac is quite nice.

[Josepi]

Bored:

264.9

My poopy 650 dollar dell:
i7/920, stock speed, thanks dell -_-
6gb 1033
4850 512
w7.64/cs4.64

[scooby24]

Do I win? :-)

151.4 seconds

i7 920 d0 @ 4.4ghz
6gb Corsair DDR3 @ 1683
Asus Rampage II extreme
GTX 295
Window 7 64/cs5 64 bit

Picture: http://i79.photobucket.com/albums/j123/ga9213/IMG_0354.jpg

[iHD]

Just gave it a try!

Everything as its stock speed

i7 930 (2.80)
6Gb G.Skill DDR3 12800
ASUS P6X58D-E & ATi 5770 GDDR5
Win7 64bit & CS5 64 bit

and the score was 223.7!

I don't know should I be happy or what!? ;)

[james_in_baltimore]

Well, I'm not doing so bad with my lowly quad core. It was well worth the upgrade

W7-64 and CS4-64
core 2 quad q9400 @ 3.4ghz (overclocked from 2.66)
Asus P5E-VM-HDMI
8GB DDR2 at 850mhz and 5-5-5-15 timings
500GB caviar green
ATI 5570 1GB

226.2 seconds

Not too bad for a 3 year old computer that was recently upgraded. I bet a hard drive upgrade would help even more.

[Bobster]

running on my Quad Core Q8400 @ 2.96GHz

276.7 Sec, a 43.2 second improvement for a small outlay after selling my Dual Core CPU :)

[homera1]

I found this test really useful. I did it on my old pc and it was rubbish, but just built my new PC and got 159.8secs

i7 930 d0 @ 4.11ghz H50 cooling with 2xCoolermaster Excalibur in Push/pull
6gb Corsair DDR3 @ 1571
Asus P6x58D Premium
GTS250
Window 7 64/cs5 64 bit

[snyper77]

Just upgraded my PC:

W7 64bit Evaluation Copy: 7100

i7 920 OC'd to 3.4Ghz on stock cooler
6GB Patriot DDR3 1600Mhz RAM
Asus P6T Deluxe V2 Motherboard
1TB 7.2K 32MB WD Caviar Black HDD
Corsair 650W TX PSU
1x Nvidia GTX260 GPU
Coolermaster Sniper Case
CS3 Extended (I think it would be significantly quicker in CS4 sine this would utilise the GPU better)

234 seconds flat. (Sure beats the hell out of my old Core 2 Duo system!)

I got similar results (and a similar setup as yours)
229.8 seconds here

i7 920 at 2.6ghz
6gb ram
1tb samsung hd
1gb video radeon 4850
Win7-64
CS4-64
coolermaster HAF922 case

[bigpow]

> compatible from PS7 up to CS4

Wish they had come up with a synthetic PS benchmark instead, so PS doesn't have to be installed

[Drakeskakes]

New macbook air 1.8gz 4gb ram SSD and I got 373
Ill run my i7 in the morning

[Drakeskakes]

i7 860
4gig DD3
ATi 5760 HD
Wd 7200 RPM HDD 1TB
Windows 7 x64
CS5

281.0
Is that bad? I feel like most i7's are running faster with the same gb ram?

[soban]

i7 920 @ 4.2 GHz
6GB PC3 12800
HD 5870
OCZ Vertex 2 40GB
Photoshop CS5

Total time: 158.9

[Drakeskakes]

why does everyone have 920s..?

I read the 860 was more suited for photoshop and the 920 for gaming and video.... was I informed wrong?

[Bobster]

i7 920 @ 4.2 GHz
6GB PC3 12800
HD 5870
OCZ Vertex 2 40GB
Photoshop CS5

Total time: 158.9

what you cooling ur rig with?

[Bobster]

why does everyone have 920s..?

I read the 860 was more suited for photoshop and the 920 for gaming and video.... was I informed wrong?

looks like there's only seconds in benchmark tests between the 2, i guess it comes down to if you want to OC and get the most out of the system to which you go for?

[Down_Shift]

i7 950 @ 4.4ghz OC, 2x 460GTX SLI
139.7 sec
Fresh install as of yesterday with 2 san controller SSD's running win7 and CS5 x64 on the SSD's.

Q6600 @ 3.4ghz , 1x 460GTX 1gb Superclocked
250.2 sec <----!!!! WHOA didn't expect this to be so high! I actually set CS5 to take up 6.5gb of ram instead of the default (4gb)... it was actually using a huge chunk of the ram.. about ~5 during filters..

[twofruitz]

Core i7 920 d0 step at 3.67ghz - Zalman 9900 cooler.
12gb Corsair DDR3 1600 RAM
Nvidia GTX470 (currently stock speeds, however voltage has been turned up)
128gb Crucial C300 SSD.
Gigabyte X58A-UD3R.

188 seconds. Seems on par with others using the same system :)

[twofruitz]

why does everyone have 920s..?

I read the 860 was more suited for photoshop and the 920 for gaming and video.... was I informed wrong?

Haven't heard that one before. 920 has HUGE overclocking potential, so it's a great choice for most.

[sempaidavid]

I can't figure out where to download the test file. I downloaded the action, but don't see a link to the file. Someone Help!

[digitaljay]

The test file should be in the DHPSV3BENCH folder after unzipping the download. It's labeled DHTestimage.jpg.

[digitaljay]

I'm pretty happy with my result of 188.2 sec. The C2Q still has some life left!

Q9550 at 3.6GHz
4x2gb G-Skill DDR-2 at 900MHz
300gb V-Raptor
Gigabyte P45-UDP3
Mac OS 10.6

[sempaidavid]

The test file should be in the DHPSV3BENCH folder after unzipping the download. It's labeled DHTestimage.jpg.


Now i see it....Thanks.
Well I tested my current Core2 Duo. I turned in a sad 284sec. I'll be upgrading next month so I wanted to do a before and after.

[Titus213]

iMac i7 2.8
2 x 2 gig Samsung
1 TB Seagate
ATI Radeon HD 4850 512 Meg
Mac OS 10.6.5

193.1 sec

2 x 4 gig memory on order. Will retest when installed.

[coleygm]

Win 7 Ultimate 64-bit
I 980x @ 3.4 GHz
12GB Corsair DominatorGT DDR3-16000
128GB Crucial C300 SSD SATAIII
NVidia GTX480 1.5GB
All Air Cooled
CS5
191.4 sec

This will be my benchmark, as I'll be adding a Velociraptor RAID 0 array for Scratch Disks and OC'ing everything shortly. I really wonder how much a difference it'll make though.

[Tony_A]

A much more relevant benchmark for photographers than the almost worthless Windows Experience rating. I have to download this & give it a try.

[Sp1207]

why does everyone have 920s..?

I read the 860 was more suited for photoshop and the 920 for gaming and video.... was I informed wrong?

Actually the 860 is a theoretically faster chip for just about everything but database work/throughput bound photoshop work (like this test). It's also the 'better' overclocker.

[sempaidavid]

My new HP elite HPE 390t just turned in a time of 190 sec. Too bad I can't OC with this motherboard.

[solara]

i7-860 2.8 Ghz
12 gb DDR3 1333 Mhz
nVidia GT220 1gb
OCZ Vertex 2 60 gb
Win7 64-bit Home Premium
PS CS-5 64-bit

Total time: 210.8 sec

[haroldwilson]

201.5 seconds

Built Christmas Eve:
I7 950 stock speed (for now)
12 G DDR3/1333
P6x58D
Intel SSD for cache
Intel SSD for OS/Apps
CS5 64 bit

I tried this with the Basic, Normal and Advance setting in the GPU settings under Performance in PS. I also set the memory slider at 8g and at 12g. None of these things made a noticeable difference. They were all plus or minus .4 seconds.

[peat]

tried this on my 3 yr old pc I built. Just to see if if I really needed to splash out on an upgrade.
cpu = intel quad core q9450 @ stock
mem = 2x2gb ocz reaperx 1333 @ stock
hd = standard sata normal thing.
win 7 ultimate 64bit
cs5

277.6

I then added some old 2x2gb ocz platinum slower memory to it to give it 8gb did the test again and got 261.0

Quite pleased with that for such an old system although it now means I cant justify buying and building a brand new pc... doh. :cry:

[coleygm]

Win 7 Ultimate 64-bit
I 980x @ 3.4 GHz
12GB Corsair DominatorGT DDR3-16000
128GB Crucial C300 SSD SATAIII
NVidia GTX480 1.5GB
All Air Cooled
CS5
191.4 sec

This will be my benchmark, as I'll be adding a Velociraptor RAID 0 array for Scratch Disks and OC'ing everything shortly. I really wonder how much a difference it'll make though.

OverClockced to 4.3GHz now with a new time of 154.3 sec

Now...i wonder will it make a difference when i add proper scratch disks???

[lens pirate]

OverClockced to 4.3GHz now with a new time of 154.3 sec

Now...i wonder will it make a difference when i add proper scratch disks???

Try adding a compressed 4 gig ram disk and making THAT your photoshop scratch disk.

That is if you really want fast. That would be about 250 to 500 times faster that a raid 0 scratch disk. It would also give some of your RAM a actual useful task.

[sempaidavid]

I just added a 4gb ram disk and set it as the scratch disk. My time of 191 sec. did not improve. I would have thought that this would have made a difference. I guess something else is limiting my speed perhaps.

[coleygm]

I just added a 4gb ram disk and set it as the scratch disk. My time of 191 sec. did not improve. I would have thought that this would have made a difference. I guess something else is limiting my speed perhaps.

Until RAM Disks are native to the OS (someone correct me if they now are), i'm avoiding them. ...plus i've read on countless PS forums as to issues and/or no performance inprovement when using RAM disks.

Someone please educate me if you think i'm wrong on this though.

[mds2004]

Not sure what took mine so long...

i7 950 @ 4GHz
12GB 1450MHz
OSX 10.6.6
CS5 64bit


340.2 seconds...

HD is a 2Tb 7200rpm, but partitioned to 2x 1TB

[coleygm]

Not sure what took mine so long...
i7 950 @ 4GHz
12GB 1450MHz
OSX 10.6.6
CS5 64bit
340.2 seconds...
HD is a 2Tb 7200rpm, but partitioned to 2x 1TB

The partitioned HD isn't doing you any favors from a perfromance perspective, but would agree that 340sec's seems hight for the hardware you mentioned....then i saw the OSX part. ;)

Maybe do check your PS performance settings though...maybe your PS settings are causing the slow time.

[mds2004]

The partitioned HD isn't doing you any favors from a perfromance perspective, but would agree that 340sec's seems hight for the hardware you mentioned....then i saw the OSX part. ;)

Maybe do check your PS performance settings though...maybe your PS settings are causing the slow time.

Will do. I also plan on buying an SSD soon as I am about to get some Amazon credit via GiftCardRescue.

Is OSX usually slower than WIndows 7? Mine also may be slower because it is a Hackintosh versus a legit Mac?

[coleygm]

Will do. I also plan on buying an SSD soon as I am about to get some Amazon credit via GiftCardRescue.

Is OSX usually slower than Windows 7? Mine also may be slower because it is a Hackintosh versus a legit Mac?

The SSD's are incredible from a Read perspective, but don't do anything for you in the read area...especially vs. a RAID 0 array. So ideally you'll want the PS program on the SSD, but keep your data files and scratch disks on traditional. Also, not all SSD's are created equal...so pay attention to the read/write speeds when your shopping. Cheaper SSD's are often no better than a traditional HD.

Regarding the Mac, I'll go with a 'No Comment' as it will surely bring a thread ending onslaught from the Apple Cult. You can look up current performance and benchmarking results that are readily available on the net, and make your own decision. :D

[link604]

436.7

AMD ATHLON 64 X2 5600+
3G RAM
Integrated NVIDEO GEFORCE 5130

[Asquared]

I don't have a scratch disk set up with PS yet.
But I ran it anyway. I got 186.8.
mac pro
3.2ghz Quad
16GB DDR3

[fetish]

yes i registered just to shamelessly post my score after i installed my new ram this morning. =p

i7 920 oced to 4ghz
24gb gskill pc16000 cl9
4x 7200.11 320gb seagate on raid0
nvidia gtx295
win7 ult 64bit
ps cs5 64bit

managed 159sec

but strangely a 4gb ramdisk as scratchdisk didnt make a single second of difference.

[Johnny9s]

thought i would try this with my new build

i5 2500k sandybridge auto overclock to 4.3 on air
8 gb gskill ddr3 1333
gigabyte gtx 570
WD caviar black 640gb
windows 7 ultra 64bit
ps cs5 64bit

149.9 seconds

[thegrandpoohbah]

Decided to use this to benchmark my new system build

- Intel Core i7 2600K
- ASUS P8P67 Deluxe
- Corsair Cooling Air Series A50 CPU Cooler
- Corsair Vengeance Blue 8GB 2X4GB DDR3-1600 CL9-9-9-24 Dual Channel Memory
- Gigabyte Radeon HD 6850 Windforce 820MHZ 1GB 4.2GHZ Video Card
- Kingston SSDNow V100 Series 128GB 2.5IN SSD
- Western Digital Caviar Black 1TB x 2
- Windows 7 Home Premium 64 bit

Result: 171.3 seconds.

[Damian75]

Just for the hell of it I tested my old MBP early 2008

2.4 GHz Core 2 Duo
4 GB 667 DDR2
200 GB 5400rpm HD
OSX 10.6.7

Results : 367.5
Not too bad for an old laptop with a slow drive in it.

[Bobster]

3Ghz Dual Core - 8GB RAM, Win7 64

usual shizzle running - CaptureOne in the background processing some images from yesterdays wedding, Winamp blaring, Firefox with this page and a few extras running..

319.9 Seconds
nothing but Photoshop running
265.8 sec - just doing this before running on new system tomorrow once i've built it.

[Bobster]

new rig
i5 2500K @ 3.3GHz (Stock until my H60 arrives) - 16GB DDR3 1600MHz Corsair Vengeance
169.3 sec

[Johnny9s]

Just overclocked my Sandybridge i5 2500k to 4.9ghz on air. Ran this test and CPU never got above 44 degrees.

i5 2500k sandybridge overclock to 4.9 on air
8 gb gskill ddr3 1333
gigabyte gtx 570
WD caviar black 640gb
windows 7 ultra 64bit
ps cs5 64bit

123.9 seconds (took 26 seconds off from gentle overclock at 4.3)

[lettershop]

HP H8-1010 with i5-2370T, Windows 7 64 Bit, 8 GBytes of EAM

156.1 seconds

[Going Baroque]

Been feeling the urge to upgrade after a minor episode with a blue screen of death (it was just the memory that went bad), so I thought I'd run the test to see if it's worth it.

262.4

E8500 @ 3.16 Core 2 Duo
4 GB DDR3
Windows XP :D
Photoshop cs4

I was considering dumping about $550 into the upgrade (i7 950 + ASUS mb combo on sale at frys for $350, and max out the ram at 24 GB for around $200 only because it's just a little more than 8-12 GB), but I'm not so sure anymore after seeing some of the numbers in this thread. I spend a lot of time in Lightroom because I do event photography, and my main complaint is the little bit of lag when I go from photo to photo. For what I do, I have a feeling I would get more bang for the buck if I just clean up my libraries (though they're not that bad), make use of a separate scratch disk and go to 8 gigs of ram on my current rig. Thoughts?

Oh, I should add that I just picked up a 5dm2 and have been playing with videos on Premiere. I won't be doing much with video (more of a novelty to me right now), but I'd like to mess with clips every now and then.

[Curtiss Bryant]

This doesnt make sense to me.. I am looking at systems that should be slower than mine, yet they are beating my times pretty significantly. I just ran it twice and got 276.9. I would think my system should get it done faster than that.

AMD Phenom II 955 @ stock 3.2ghz
8gb ram
64 gb SSD
Geforce 9500gt

I would think it should be able to beat a core2 laptop.

[Bobster]

old system (ran just before dismantling) Q8600 @ 2.96GHz - 265.8
new system (before overclocking) i5 2500K @ 3.3GHz - 169.3
new system (overclocked) i5 2500K @ 4.3GHz - 135.4

(system didn't go above 27% CPU usage the whole time)

[Chris1le]

Asus Essentio 6850-07
I7 2600 Stock
8GB RAM
WD Black 1TBx2
Win7 64bit
PS CS5.1 64Bit

162.7

[tim]

i7-2600K, 16GB RAM, 2xOWC SSD as scratch. Win7-64, CS4-64. 161.2s

Seems the test is mostly single threaded. I looked at CPU usage at one point and it was 12.5% or so. My real world workloads can use multiple threads.

[Revo]

162.3s for me. It didn't seem to make effective use of my processor. It never got near 100% load.

Here is my setup:

Asus P8P67 Deluxe
i7-2600K at stock clock (have had it overclocked stably at 4.5ghz, but since I never seem to max it out at stock speeds I just left it there until I run into a situation where I need more processing power)
8GB Ram at 1600mhz and 7-8-7-24-1 timing
GTX 580 1.5GB Video Card (I game with this computer as well)
240GB PCIexpress SSD (for OS, programs, and scratch - 1500mb/s read, 1225mb/s write)
Two 2TB WD Black in Raid 1 for storage purposes.
PS CS5
Windows 7 Pro 64

[turbo_911]

229s for me and seems a bit slow for my set-up.

AMD Phenom II X4 965 @3.6Ghz
8GB G.Skill DDR1600 @ 1721Mhz and 7-7-7-24 timings
GTX 460 OC 768MB
90GB SSD Performance 3
PS CS5
Windows 7 Ultimate 64

[CAReefer]

139.5 On a fresh build/upgrade

Asus P8Z68V Pro
Core i7 2600K at stock 3.4 GHz
16 Gb Corsair Vengeance 1600 running in XMP
OCZ Vertex 3 120 Gb SSD
Asus GTX 560 (fermi) 1 Gb DDR5
PS CS5
Windows 7 Premium

[Dooms_day]

Still waiting for GPU support so my GTX 560 ti can take it away... Getting around 300 like everyone else.

[Talley]

138.8 seconds is this the fastest yet?

Asus P8Z68-V/Gen 3
Intel 2500K @ 4.3Ghz w/ Cooler master hyper 212+
8GB Corsair Vengence DDR3 1600
Crucial M4 64GB SSD drive
(2) Seagate 500gb 7200 in raid 0 for data drive
NO video card yet....

I should put this chip back to 4.6ghz @ 1.38v to see what I can do but I like my 4.3 @ 1.28v best.

[digitaljay]

I managed 131.5 seconds on my new build.

Intel 2600K @ 4.6Ghz
Gigabyte UD3H-B3
16GB Kingston DDR3 1600
Kingston 128 GB Hyper-x SSD
2x WD RE3 1TB raid 0 for data
OCZ 64 GB Agility SSD for scratch
XFX 6870 vga
Mac OS X 10.7

[Talley]

I managed 131.5 seconds on my new build.

Intel 2600K @ 4.6Ghz
Gigabyte UD3H-B3
16GB Kingston DDR3 1600
Kingston 128 GB Hyper-x SSD
2x WD RE3 1TB raid 0 for data
OCZ 64 GB Agility SSD for scratch
XFX 6870 vga
Mac OS X 10.7
I'm glad I saved money. The extra money I would've spent just didn't justify it for me considering the bang for the buck of mine. Thanks for the post.

[jrader]

Intel i7-2600K CPU @ 3.40GHz, 4 Cores
16GB RAM DDR3 1600
2x1GB Graphics card (not that this matters to the test)
Windows 7 Home Premium
PS CS5x64

161.9 seconds

I allotted all available RAM to PS (13.9GB) and used a 1.5TB hard drive as the scratch drive. I set the history states to 1 and the cache to 8. I am confused by the following though. I had the task manager resource monitor visible while running this test and during the test, I witnessed that the CPU Usage was never above 12% and the Memory usage was never above 2.75 GB.

Why doesn't PS utilize all (or more of the) available resources, especially when I'm letting it? I feel like it should run much faster without having to overclock my processors (which I can, but not willing to quite yet).

Insight anyone?

John

[coleygm]

Intel i7-2600K CPU @ 3.40GHz, 4 Cores

16GB RAM DDR3 1600

2x1GB Graphics card (not that this matters to the test)

Windows 7 Home Premium

PS CS5x64



161.9 seconds



I allotted all available RAM to PS (13.9GB) and used a 1.5TB hard drive as the scratch drive. I set the history states to 1 and the cache to 8. I am confused by the following though. I had the task manager resource monitor visible while running this test and during the test, I witnessed that the CPU Usage was never above 12% and the Memory usage was never above 2.75 GB.



Why doesn't PS utilize all (or more of the) available resources, especially when I'm letting it? I feel like it should run much faster without having to overclock my processors (which I can, but not willing to quite yet).



Insight anyone?



John


Biggest issue with these tests is exactly what you mentioned. Depending on the feature, filter, effect, etc... some are multi core/thread capable and others are not. I had 4 yr old Quad Core that would come close to what my new i7 monster rig would report as far as Photoshop speed tests, so don't put to much stock into it. Not to mention anyone can manipulate performance settings just for this test to give themselves lower scores...but that would have nothing to do with real-word or normal Photoshop usage or needs unless those needs were the exact events the speed test was performing. I had 4 yr old Quad Core that would come close to what my new i7 monster rig would report as far as Photoshop speed tests, so don't put too much stock into it.

I'm sure in future releases (ie CS6 and beyond) we'll start seeing usage of multi-cores more universal though hopefully. Currently where you really see the advantages of the big multicore rigs is in the video processing software such as Adobe Premiere Pro. Now that has been a night and day difference from my old machine, as it uses every bit processor you have. Had to replace my air cooler with a water cooler just to keep my core temperatures reasonable when encoding.

[paulo.maxim]

143,1 Seconds
My PC:
Intel i5 2500
8gb ram ddr3 1600
Ati 6950
Asus Maximus Formula IV
SSD 128 Samsung
Windows 7 64bits
CS5 x64

[Pit]

The operations required do not warrant more ram usage. As for CPU, your disk might have been the bottleneck, Was it 12% on each core or 12% combined aka 1 core at 50% and 3 at 0%?
I'll do some tests with this when i get home, my work machine is an old core2quad based xeon Mac Pro with 2 quad core cpus.

161.9 seconds


I allotted all available RAM to PS (13.9GB) and used a 1.5TB hard drive as the scratch drive. I set the history states to 1 and the cache to 8. I am confused by the following though. I had the task manager resource monitor visible while running this test and during the test, I witnessed that the CPU Usage was never above 12% and the Memory usage was never above 2.75 GB.

Why doesn't PS utilize all (or more of the) available resources, especially when I'm letting it? I feel like it should run much faster without having to overclock my processors (which I can, but not willing to quite yet).

Insight anyone?

[jrader]

The operations required do not warrant more ram usage. As for CPU, your disk might have been the bottleneck, Was it 12% on each core or 12% combined aka 1 core at 50% and 3 at 0%?
I'll do some tests with this when i get home, my work machine is an old core2quad based xeon Mac Pro with 2 quad core cpus.
I think it was 12% combined.

[Pit]

Yeah Photoshop is probably not multithredding most filters and actions. So at some time you can have 1 thread out of 8 pegged at 100% while others sit idle (theoretical scenario or 4core 8thread intel 2600k chip). I'll take a closer look at my machine once i get home.

[Pit]

124.8 FIrst go around, not clean after restart.
123.2 After restart.

i7 2600k @4800mhz
16g 1600mhz
240g ocz agility 3 ssd
2x 6970 2g gpus

[Sfordphoto]

226 sec

i7-860 2.80GHz stock
Gigabyte P55A-UD4P
8GB RAM 1333MHz
128GB Crucial M4 SSD
GTX 460 1GB
Win7 64-bit
PS CS5 64-bit