'Real World' Photoshop Benchmark - Speed Test

[GXPvince]

Very surprised with my 1,1 - was thinking more like high 20's

Mac Pro 1,1
2.66 Quad x 2 (8 core)
16gb RAM
ATI 5770
2-128 SSDs (1 for cache 1 for apps)

22.9 22.8 22.6

[TheEasterBunny]

This test helps me a lot.
It helped get my Photoshop tuned in nicely.
Specs:
1,1 Mac Pro
4GB RAM
Dual 5150's dual core@2.66Ghz (total 4 cores)
HD:
2) 128GB SSD's in a RAID0 array (one of the scratch disks)
1) 250GB 7200 RPM (one of the scratch disks)
Video Card:
NVIDIA GeForce 7300 GT
CS5 Settings:
History states 1
Cache Levels 4
RAM for PS at 3800MB

Time: 27s

[Josh-H]

16.4 Seconds

Late 2008 Mac Pro 8 Core
32 Gig of Ram
PS CS6
4 x WD RED Drives in RAID 10
Nvidia Geforce 8800 GT 512MB
History States 1
Cache Levels 4
RAM for PS MAX

[jmacpo]

Hey Guys: If you have read any of my previous posts regarding my "Search for Speed" I'm going to take this opportunity to share what I've found out with my usb 3.0 purchases.
Since I'm a Photographer working with large Raw files requiring multiple backups etc. I was drawn to the claims of the usb 3.0 interface! Knowing CalDigits reputation I took the plunge and purchased the VR2 along with their 3.0/esata PCIe card with hopes that this would be the answer. Not that I expected 5Gb/s(625MB/s) I did expect to achieve 180-200MB/s on my backups! Well even those numbers were not achievable on some of the best drives out there such as the Hitachi 2TB 7300K and the WD Black 2TB 64MB 7200RPM!
The only way to come close to those Data Rates were with Raid 0 setup on the VR2 with my best numbers being 174MB/s Read and 167MB/s Write which I'm more than pleased with!
But I still had to contact CalDigit for an explanation as to why they would advertise 5Gb/s when they know it is not possible for anyone to come close to those numbers? See Below for their explanation-

***********************************************

Dear customer,

The VR2's USB3.0 performance is around 180MB/s ~ 210MB/s, and its read speed is higher than the write speed. If you want the maximum VR2 performance (230~250MB/s), you would need the eSATA connection.

The bottleneck is on the USB3.0 card, which it does not support UAS(http://en.wikipedia.org/wiki/USB_Attached_SCSI). However, despite the limitation, the USB3.0 card does provide a relatively close range within the VR2's maximum performance.
- None of the USB3.0 card currently support the UAS (including Lacie/Sonnet/HighTech etc…). The only UAS unlocked USB3.0 connection are the native USB3.0 ports on the 2012 Mac computers with the built-in USB3.0 ports.

Best regards,

CalDigit - Support

*******************************************
So I guess you guys with the new Macs may be the only ones on the planet with any hopes of achieving the 200 plus MB/s but the real surprise to me was their claim that one could expect even higher speeds with eSATA connections?

PHOTOSHOP TEST RESULTS: 11.2secs

My Mac Pro Workstation:
Mac Pro 3.33GHz 6 Core Intel Xeon
32Gb DDR3 1333 Memory
Accellsior PCIe 240GB SSD "Raid 0" Boot Drive [680MB/s AJA SpeedTest]
Dual 600GB WD VelociRaptor 10,000RPM SATA 6 Gb/s "Raid 0" 1.2TB
CalDigit VR2 4TB External "Raid 0" Drive

[Tesselator]

@ jmacpo:
http://www.hptmac.com/product.php?_index=102
__________________________________________________________________


All,
I thought I would revisit these benchmarks now some four and a half years later. I still use the same machine although it's now my web-browsing machine mostly. I added RAM to bring it up to 32GB and just recently replaced the 7300 NVidia card with the 8800GT. I'm of course using CS6 for these tests and the driverheaven test suite is now in version 3 which is considerably different from the version one suite used below here. Because of this the times are not relative at all but I'm quoting the message from Page 30 none the less.

I have to add that this MacPro1,1 has been the best machine ever - on the cost|performance|longevity curve. Although it's at a dead end as far as OS upgrades go with 10.7.5 being the last supported version it still keeps up with modern offerings. Even the super over-clocked 4.5GHz 6-core screamers that tweakers spend days optimizing are not twice as fast - although fairly close to twice...

This is different than machines twice this age. So it seems the progression of processor and system speed over time has been slowing down considerably! Or to put it another way: Seven years before the MacPro1,1 the fastest machines were less than or about one sixteenth (1/16) the speed of the MacPro1,1 and now seven years after the MacPro the fastest machines are not even two times faster than the MacPro1,1 (or if a few are then not by very much - and those will be WAY WAY above the MacPro1,1's introductory price-point!). So what the hell happened to More's Law? Heh, some law... Right after it was made up it failed itself.

So anyway here are the current results (and I might add that CS6 is about 10% slower than CS4 at most things on this hardware):

CS6 || MacPro 1.1 || X5355 2.66 8-Core || 32Gigs RAM || No ZDNet's OverClocker

Performing the PhotoShop benchmark from this site: http://www.retouchartists.com/pages/speedtest.html The results are:

• At 2.66 GHz - 22.1 Seconds

Performing the PhotoShop benchmark from this site: http://www.driverheaven.net/photoshop.php:

• At only 2.66GHz (now Version 3 of the test suite):
  Texturiser:............... 1.1s
  CMYK:..................... 0.8s
  RGB:....................... 1.0s
  Ink Outlines:............ 15.3s
  Dust & Stratches:..... 1.3s
  Watercolor:............. 14.3s
  Texturiser:.............. 1.4s
  Stained Glass:......... 12.6s
  Lighting:................. 2.9s
  Mosiac:.................. 10.0
  Extrude:................. 79.5s
  Smart Blur:............. 41.4s
  Underpainting:......... 20.0s
  Palette:.................. 15.5s
  Sponge:................. 15.7
  
  Total Score:..............232.8 sec.
  

Originally Posted by Tesselator

Here's my specs & results:

CS4 || MacPro 1.1 || X5355 2.66 8-Core || 4Gigs RAM || ZDNet's OverClocker

Performing the PhotoShop benchmark from this site: http://www.driverheaven.net/photoshop.php at two CPU speeds:

• At 2.66 GHz
  Texturizer:............... 1.03 sec.
  CYMK CC:................. 1.27 sec.
  RGB CC:................... 1.54 sec.
  Dust & Scratches:...... 1.63 sec.
  Water Color:............. 11.32 sec.
  Texturizer 2:............. 1.21 sec.
  Stained Glass:........... 3.36 sec.
  Lighting Effects:........ 3.41 sec.
  Mosiac Tiles:............ 10.56 sec.
  Extrude:.................. 44.52 sec.
  Smart Blur:............. 62.85 sec.
  Underpainting:......... 13.92 sec.
  
  Total Score:..............156.62 sec.
  
  
• At 3.06 GHz
  Total Score:..............138.81 sec.

Performing the PhotoShop benchmark from this site: http://www.retouchartists.com/pages/speedtest.html The results are:

• At 2.66 GHz - 28.53 Seconds
• At 2.76 GHz - 27.43 Seconds
• At 2.86 GHz - 26.35 Seconds
• At 2.93 GHz - 25.32 Seconds

• At 2.97 GHz - 25.53 Seconds
• At 3.01 GHz - 25.08 Seconds
• At 3.10 GHz - 25.53 Seconds

Me thinks PhotoShop is not very good for profiling CPUs. The speed bar in ZDNet OC changed from green to yellow at 2.968 GHz (2.97 GHz) thus the colors in the list above. I guess farther up it turns red but I didn't go there!

[mtasquared]

15.5 seconds

Mac Pro 5,1
Quad-Core 2.8 GHz
10 GB ram
180 GB Intel SSD, boot
1 TB SATA disk
Radeon 5770
CS6

Shaved off more than 2 seconds with tweaks, and using CS6. Cheers!

[Anastasia72]

1:01

2009 Imac
3.06 GHZ Intel Duo Core
12 gb ram
Nvidia GeForce 9400
running 10.8.2

Using Photoshop CS4 extended

I am so glad I am grabbing a new machine, this time is terrible

----------

Originally Posted by mtasquared

15.5 seconds

Mac Pro 5,1
Quad-Core 2.8 GHz
10 GB ram
180 GB Intel SSD, boot
1 TB SATA disk
Radeon 5770
CS6

Shaved off more than 2 seconds with tweaks, and using CS6. Cheers!

you just confirmed my choice for the 6 core Good time there!

[Tesselator]

I've been continually vindicated over the years going with more cores rather than fewer and faster. I do wish these could be overclocked or were available for a reasonable price in 4.0 or 4.5GHz tho - fast AND many would of course be even better. IMO there's almost no difference between 2.66 and 3.0 or between 2.8 and 3.2 though - I doubt I could ever notice the difference even if I wanted to.

And 2.66/2.8GHz is a nice sweet spot. Low power con$umption while I'm reading, 8 cores available when I need the power potentially delivering 21.28GHz or 22.4GHz respectively, and a nice low system temperature!

Had I to choose again I would definitely choose 6 over 3, 8 over 4, or 12 over 6 cores! People are still often seen claiming that most software doesn't offer multi-proc support but I've found that just about everything I would want to support MP indeed does.

[luisramos]

13.3 seconds

Mac Pro 2012
12-core 2.4GHz
24 GB ram
120 GB Intel SSD, boot
1 TB SATA disk
Radeon 5770
CS6

[zek0]

Originally Posted by zek0

10.2 sec - CS5

Mac Pro 2x2.93 12 Cores Radeon 5870 48GB 2xOWC SSD PRO RE Raid 0

CS6
Upgrade Nvidia GeForce GTX 670

9.4

[jmacpo]

Just thought I'd add my 2cents:

RE: DriverHaven's SpeedTest - Total Time of 163sec

RE: RetouchArtist SpeedTest - Total Time of 11.2sec

I'm Happy

Running on Photoshop CS6

My System: MacPro 5,1
Processor- 3.33 GHz 6-Core Intel Xeon
Memory- 32 GB 1333 MHz DDR3 ECC
Graphics- ATI Radeon HD 5870 1024MB
Software- Mac OS X Lion 10.7.5

[benwiggy]

I have a 2012 MacMini, 2.6 i7, 16Gb RAM, Fusion drive, running Creative Suite 6.

13 seconds.

I just loaded the image and ran the RetouchArtist Speed Test action. I didn't change any settings or close other apps. Sorry if that renders the test incomparable.

It's still a pretty staggering result, I'd say.

[mseth]

Ran the test and got a time of 16.4s. Using CS6, system is all up to date.

System:
Mac Pro 4,1 - 2.66GHz QC
16GB RAM
128GB SSD
GTX 670

(see sig for more details)

[Killerbob]

Finally got around running this test on my PS CS6: 16,8 seconds

Mac Pro 2008
2 x 2.8GHz Quad-core
16GB RAM
960GB Accelsior card for boot and apps
eVGA GTX680 SC

[Callumbear845]

I am really impressed with how powerful modern laptops have become. My 2012 rMBP was able to do the test in 20.1 seconds while I was watching hulu on Chrome, a half dozen tabs open in safari, aperture open, a very large excel sheet open, iTunes open, a slow HDD with 5 partitions plugged and a few word documents open.
After a restart, it did it in only 15.2 seconds!

Specs:
2012 15" rMBP
2.3 GHz Quad Core i7
16GB 1600MHz RAM
256GB Samsung 830
nVidia 650GT w/ 1GB vRAM
Photoshop CS6 Extended

[Macsonic]

Originally Posted by Tesselator

I've been continually vindicated over the years going with more cores rather than fewer and faster. I do wish these could be overclocked or were available for a reasonable price in 4.0 or 4.5GHz tho - fast AND many would of course be even better. IMO there's almost no difference between 2.66 and 3.0 or between 2.8 and 3.2 though - I doubt I could ever notice the difference even if I wanted to.

And 2.66/2.8GHz is a nice sweet spot. Low power con$umption while I'm reading, 8 cores available when I need the power potentially delivering 21.28GHz or 22.4GHz respectively, and a nice low system temperature!

Had I to choose again I would definitely choose 6 over 3, 8 over 4, or 12 over 6 cores! People are still often seen claiming that most software doesn't offer multi-proc support but I've found that just about everything I would want to support MP indeed does.

I guess you may be correct. I am using "Menu Meters" to measure CPU usage in my 2008 8 core Mac Pro. Just noticed in using applications that are single thread, all 8 cores go to work from what I see at the Menu Meter. Specailly with Photoshop, which only uses maximum of 2 cores from what I read, like a simple image rotate or image resize, all 8 cores are busy working and not a single core that is idle. Not sure how this is happening. My friends tell me it's a waste to buy a multi-core Mac Pro if you're just using single threaded apps.

[GXPvince]

Originally Posted by Macsonic

I guess you may be correct. I am using "Menu Meters" to measure CPU usage in my 2008 8 core Mac Pro. Just noticed in using applications that are single thread, all 8 cores go to work from what I see at the Menu Meter. Specailly with Photoshop, which only uses maximum of 2 cores from what I read, like a simple image rotate or image resize, all 8 cores are busy working and not a single core that is idle. Not sure how this is happening. My friends tell me it's a waste to buy a multi-core Mac Pro if you're just using single threaded apps.

I have been using my 12 core for a few days now and this is what I experienced. PS does at some points use all cores specifically some actions I use. However, if you look at the % used its no where near maximum.
Next, I was able to use about 16 cores when exporting from media encoder and exporting pictures at the same time from lightroom. I only have 12 GBs of ram right now so that might have been the limiting factor, especially on the media encoder export.

[Macsonic]

Originally Posted by GXPvince

I have been using my 12 core for a few days now and this is what I experienced. PS does at some points use all cores specifically some actions I use. However, if you look at the % used its no where near maximum.
Next, I was able to use about 16 cores when exporting from media encoder and exporting pictures at the same time from lightroom. I only have 12 GBs of ram right now so that might have been the limiting factor, especially on the media encoder export.

Yep all cores are not used at maximum levels. In my opinion, though I am not a technical expert, this is just fine that the cpus still has extra resources rather than being maxed out or "stressed" to its max. i was chatting with a Mac tech guy as sometimes they get repairs for Mac Pros where the cpus died. Though there is no definite answer why some cpus died, his guess is could be the cpus are always stressed to the max. And when cpus are maxed out tends to bring temperatures up.

[echoout]

11.5 seconds

Mid-2012 MacBook Pro (non-Retina)
2.7Ghz Intel Core i7
16GB DDR2 1600Mhz RAM
NVIDIA GeForce GT 650M 1024 MB
2x 480GB OWC Mercury EXTREME Pro 6G SSD
OSX 10.8.3
PHotoshop CS6

I'm hoping for a MacPro to come along and beat this score handily! And sooner than later.

[GermanyChris]

Originally Posted by echoout

11.5 seconds

Mid-2012 MacBook Pro (non-Retina)
2.7Ghz Intel Core i7
16GB DDR2 1600Mhz RAM
NVIDIA GeForce GT 650M 1024 MB
2x 480GB OWC Mercury EXTREME Pro 6G SSD
OSX 10.8.3
PHotoshop CS6

I'm hoping for a MacPro to come along and beat this score handily! And sooner than later.

This test is write intensive you have SATAIII..that would be the difference between your 11 seconds, and most modern Mac's 15 or so. My sandy hack was 9 seconds or so..

[echoout]

Originally Posted by GermanyChris

This test is write intensive you have SATAIII..that would be the difference between your 11 seconds, and most modern Mac's 15 or so. My sandy hack was 9 seconds or so..

Exactly. I would hope that would be the maximum to justify getting anything new. If I'm going to throw down on a desktop it needs to really whip my laptop on things besides just rendering and encoding.

[GermanyChris]

Originally Posted by echoout

Exactly. I would hope that would be the maximum to justify getting anything new. If I'm going to throw down on a desktop it needs to really whip my laptop on things besides just rendering and encoding.

But it likely won't especially in Ps

[echoout]

Originally Posted by GermanyChris

But it likely won't especially in Ps

Holding out hope for AE and C4D speed improvements over the 2012 MP I just sold. Connectivity options would be good too.

[GermanyChris]

Originally Posted by echoout

Holding out hope for AE and C4D speed improvements over the 2012 MP I just sold. Connectivity options would be good too.

Core count and graphics win there..to be completely honest I've never noticed a difference in Ps between my '09 17", Sandy Hack, MP 1,1, and now 4,1 when using each independently. On CS 3 I didn't notice a difference between my Quad and the 17. I know they are all different speeds but meh..

[echoout]

Originally Posted by GermanyChris

Core count and graphics win there..to be completely honest I've never noticed a difference in Ps between my '09 17", Sandy Hack, MP 1,1, and now 4,1 when using each independently. On CS 3 I didn't notice a difference between my Quad and the 17. I know they are all different speeds but meh..

I hear you. I teach motion graphics in a lab with early 2009 8-core MPs with 16GB of 1066MHz RAM, nVidia GT 120 512MB, and 7200RPM HDD. Not nearly as fast as my setup but I don't see the difference until we get into the heavy stuff in my advanced classes. Being 10% slower or faster isn't that huge of a deal for most tasks. But when things DO get heavy, oh boy.