Haswell E benchmarks

Hello all. It's is once again time for some benchmarks with the new release. I will also include the previous X79 4930K clocked at 4.4GHz and Dual Xeon 10 Core as reference to these. I will add them as I get them done. Please let me know if you have any questions.
5960X @ 3.5GHz
64GB DDR4 2400
1TB Crucial M550 SSD
780Ti
AVCHD 4 Layer 30 Minute test
Adobe CC2014
3 Layer - 14:35 (Match Source selected)
3 Layer - 28:57 (HDTV 1080P 23.976)
4 Layer – 16:01 (Match Source selected)
4 Layer - 31:37 (HDTV 1080P 23.976)
Red 4K to DPX 4096 x 2048 24p Full Range (max bit depth) 30 seconds of media
3 Layer - 2:08
4 layer - 2:08
Red 5K 98 Frame to DPX 5K 23.976 slow motion Frame Full Range (max bit depth) 30 seconds of media
1 Layer - 2:12
Red 6K to DPX 6K (max bit depth) 20 seconds of media
1 Layer - 1:31
Red 4K to H264 4K 30 seconds of media
4 layer - :50(Match Sequence H264)
DNG 2.4K to H264 2.4K 26 seconds of media
1 Layer - :15
AE CC 2014
Red 4K to AVI Lossless 4k 30 seconds of media
1 Layer: 2:19
5960X @ 4.5GHz
64GB DDR4 2400
1TB Crucial M550 SSD
780Ti
AVCHD 4 Layer 30 Minute test
Adobe CC2014
3 Layer - 11:36 (Match Source selected)
3 Layer - 22:54 (HDTV 1080P 23.976)
4 Layer – 12:48 (Match Source selected)
4 Layer - 24:58 (HDTV 1080P 23.976)
Red 4K to DPX 4096 x 2048 24p Full Range (max bit depth) 30 seconds of media
3 Layer - 1:54
4 layer - 1:58
Red 5K 98 Frame to DPX 5K 24 Frame slow motion Frame Full Range (max bit depth) 30 seconds of media
1 Layer - 1:58
Red 5K 98 Frame to DNxHD 1080 23.978 36 OP1A Frame 30 seconds of media
1 Layer - :12
Red 5K 98 Frame to DNxHD 440X 1080P 60 frame OP1A Frame 30 seconds of media
1 Layer - :14
Red 6K to DPX 6K (max bit depth) 20 seconds of media
1 Layer - 1:21
Red 4K to H264 4K 30 seconds of media
4 layer - :49(Match Sequence H264)
DNG 2.4K to H264 2.4K 26 seconds of media
1 Layer - :13
AE CC 2014
Red 4K to AVI Lossless 4k 30 seconds of media
1 Layer: 1:59
The playback and export performance currently with CC 2014 is now relatively consistent. The CPU threading was across all 16 threads both on playback and export now. The GPU load consistently pushed up to 90 to 98% when the benchmark tests included GPU accelerated plugins and scaling of multiple layers. The overall efficiency is far better which is why I didnt put notes after each test. The 8 Core clocked at both 3.5GHz and 4.5GHz played back 4K, 5K 98 frame (both 24 and 60 frame playback), and even 6K at full resolution playback without dropping frames. The 5K playback was smooth regardless of slow motion or full motion preview setup. The increased bandwidth and speed of the ram is definitely having an impact there. The ram usage was as high as 30GB in Premiere for the testing but AE went well over 46GB on export. GPU ram usage pushed 2.5GB on the 3GB card with 4K+ media in Premiere but normally used around 1GB for 1080. I also included some request DNxHD OP1A exports from 5K media as a comparison of media timeframe to encoding time for off line. I will be testing the 6 Core 5960K after I do some testing with the ram at stock 2133.
Eric
ADK

Reference benchmarks:
4930K @ 4.4GHz
64GB DDR3 1600
1TB Crucial M550 SSD
780Ti
AVCHD 4 Layer 30 Minute test
3 Layer - 16:33 (Match Source selected)
3 Layer - 25:32 (HDTV 1080P 23.976)
4 Layer - 28:04 (HDTV 1080P 23.976)
4 Layer – 18:58 (Match Source selected)
Red 4K to DPX 4096 x 2048 24p Full Range (max bit depth) 30 seconds of media
3 Layer - 2:05
4 layer - 2:06
Realtime Playback 4K Full Res smooth without dropping frames
CPU threaded well for export after clearing cache by switching from Hardware MPE to Software MPE and back again before linking to AME. Realtime Playback threaded ideally.
Average GPU load 35%
Red 6K to DPX 6K (max bit depth) 20 seconds of media
1 Layer - 1:43
Realtime Playback 6K Full Res smooth without dropping frames.
CPU threaded well for export after clearing cache by switching from Hardware MPE to Software MPE and back again before linking to AME. Realtime Playback threaded ideally.
Average GPU load 15%
Red 4K to H264 4K 30 seconds of media
4 layer - :52 (Match Sequence H264)
CPU Threads very well on export. GPU load peaking at 99% consistantly.
DNG 2.4K to H264 2.4K 26 seconds of media
1 Layer - :21
CPU threads very well on export.
Red 4K to Cineform 4K Film Scan 1 30 seconds of media
4 Layer - 5:44
CPU Threads poorly
AE CC 2014
Red 4K to AVI Lossless 4k 30 seconds of media
1 Layer: 2:51
CPU Threads very well on Export. Ram Preview used 45GB of ram 3x at full res
2x Xeon E5 2690 V2 CPU's @ 3GHz
128GB DDR3 1600
1TB Crucial M550 SSD
780Ti
AVCHD 4 Layer 30 Minute test
3 Layer - 29:29
4 Layer – 31:21
Red 4K to DPX 4096 x 2048 24p Full Range (max bit depth) 30 seconds of media
3 Layer - 3:20
4 layer - 3:20
Realtime Playback 4K Full Res smooth without dropping frames.
Poor CPU Threaded on export but not playback
Average GPU load 50%
Max GPU load 99%
Red 6K to DPX 6K (max bit depth) 20 seconds of media
1 Layer - 2:29
Realtime Playback 6K Full Res smooth without dropping frames.
Poor CPU Threaded on export but not playback
Average GPU load 40%
Max GPU load 99%
2x 780Ti GPU's
AVCHD 4 Layer 30 Minute test
3 Layer - 29:59
4 Layer – 32:05
Red 4K to DPX 4096 x 2048 24p Full Range (max bit depth) 30 seconds of media
3 Layer - 3:39
4 layer - 3:30
Red 6K to DPX 6K (max bit depth) 20 seconds of media
1 Layer - 2:51
Red 4K to Pro Res 4444 via Cinemartin PLIN Gold 30 seconds of media
4 layer - 3:40 + 1 min to 2min Render time in Premiere
2x Xeon E5 2690 V2 CPU's @ 3GHz
128GB DDR3 1600
1TB Crucial M550 SSD
780Ti
DNG 2.4K to H264 2.4K 26 seconds of media
1 Layer - :19
CPU threads very well on export.
Red 4K to H264 4K 30 seconds of media
4 layer - 1:14
CPU Threads very well on export
Red 4K to Cineform 4K Film Scan 1 30 seconds of media
4 Layer - 5:55
CPU Threads poorly
AE CC 2014
Red 4K to AVI Lossless 4k 30 seconds of media
1 Layer: 2:32
CPU Threads very well on Export. Ram Preview used 99GB of ram at full res

Similar Messages

How to install Windows 8.1 on Haswell Mac?

Just bought the 15" Haswell Macbook Pro Retina and I'm trying to install Windows 8.1. I'm relatively new to Mac so the instructions that I found didn't really make sense to me and seemed like they were skipping steps. My school gave me the downloadable Windows 8.1 for Mac Version that says it has to be used with an external USB. Here's the steps I've followed leading up to my problem in the best way I can explain them.
1. Downloaded this file: English_windows_8_1_professional_upgrade_w64
2. Plugged in 8gb flash drive
3. Started boot camp assistant
4. Clicked 'Create a Windows 7 or later version install disk'
5. Chose this for ISO img: /Users/.../Documents/English_windows_8_1_professional_upgrade_w64/English_windo ws_8_1_professional_upgrade_w64.img
6. Says it's formatting drive
7. Prompted with this: Your bootable USB drive could not be created
Any insight on what's going on? I have a feeling it's because it's an upgrade? But what does that have to do with the USB drive? I've used 3 different flash drives and get the same message every time.
Also, I did do all the updates on the computer immediately upon opening it. Talked to some tech people and they said the update made it so that you don't need windows 7 first? I don't know...

After 3 hours I've successfully installed windows 8.1. Everything seems to be working fine but I'm too tired to try running any programs on it yet. I basically ran into most of the problems that I read about on the forums. Here's what I did since running into the problem listed.
One thing to take note of that I noticed a lot of people did because I've seen it in some install guides was to use two separate USB drives. One for the drive files and one for the ISO. I DID NOT follow those directions. I noticed many people had some sort of blank screen if they did or something like that so I'm assuming what I did bypassed that problem. So put everything on one USB. Mine was an 8gb toshiba nothing special.
8. After getting that prompt I logged out (making sure to uncheck the option to open any applications back up)
9. Logged in and restarted boot camp.
10. Let the program do it's thing clicking continue etc along the way.
11. As the computer restarts hold the Option key.
12. Now here's where things messed up for me again. I went to the windows boot and continued through putting in the key etc. when asked to choose which partition all of them were GPT servers. Honestly I have no idea what that means. I tried formatting as the option was highlighted but still didn't work.
13. Restart the computer, again holding the option key. This time load the EFI drive.
14. Resolution was different and everything was tiny but it's all the same thing that was just done before. When prompted for the partition I chose 4* or bootcamp and then everything installed successfully.
That's all I can think of so far instructions wise and hopefully this works for others who haven't found any solutions! I'm not super techy just did a lot of trial and error and determination.

Best Practice of DMS in Auto-Industry & benchmarks

Hi Experts,
what is the best practice in DMS for auto-induastry?
Also we need centralized server and cache servers at plant level. what will be the best practice? and archiving solutions?
what is benchmark for defining content and cache server size?
Regards,
Ravindra

answered outside sdn

Problem with my MSI 290x gaming 4g performance in unigine valley benchmark.

First im dutch(netherlands) so sorry for my maybe bad english.
My system is:
Asus p8z77 v deluxe mb
intel 3770k cpu
16gb ram 1866 Corsair
2x ssd 256 gb OCZ
MSI 290x gaming
win7 64bit ultimate.
When i came home with my card i removed my old 7970 and replaced it with my new card.
Put some programs up like gpu-z cpu-z realtemp ext.
Then run Unigine-valley Benchmark with OC tool MSI at 1040 option.
My cpu at 4.2ghz OC
Score Unigine valley v1.0: 61.1 fps-score2558-min fps 30.3/max fps 112.8
Custom settings at 1920-1080 and ultra 8x aa windowed.
Seems ok result.
Now to problem i have.
So next i did (dumb me hehe) i try OC my CPU with All-suit II from ASUS to OC cpu extreme mode which stress test untill stable clock is found.
Got a crash bluescreen and then after few attemps stable 4.3.
Then did new benchmark and my score was alot lower and nomatter what i try my 290x did not pass 60fps any more and score dropped to 2100?
3dmark firestrike gave first time score of 9655 and after my disaster with cpu failed attemp 9200.
My card is stuck at 60fps max it won't get any higher?
The GPU-Z also show PCI-E 3.0 X16@ X1 1.1and when i activate render i see it only change to x16 1.1 i never see gen 3.0.
CPU-Z info motherboard also only pci-express link with x1 and x16
My questions is how can o solve this problem it seems my motherbaord don't reconize my videocard anymore?
Also posible that something with the OC attemp CPU broke something?
I also try change from AUTO to GEN3 in bios but then i get a blackscreen when i boot up it stays backscreen, have to change it switch videocards so i can see my boot again and go into BIOS.
Im at a lost here, hope one of you know a solution or whats my problem?
Hope ive supply enough info im bit of a newbie at this sorry for that.
Thanks in advance.

>>Clear CMOS<< of your board and retry.

Macbook Air Haswell-doesn't wake up

As mentioned in the title, literally, my Macbook Air doesn't wake up occasionally.
Sometimes (3 or 4 times out of 3-50) when i close the lid of my notebook and open, nothing appears but the black screen appears and only the light of the keyboard glows. No input(including trackpad and the keyboard) helps get out of the black screen and the only thing i should do is to turn the power of and turn it on again.
It seems that some users using Macbook Air Haswell and some Macbook Pro models. And some explains it happens because of chrome browser. I purchased this notebook about a week ago and I can get a new one by contacting Apple Store. Should I get a new one and see if it happens? or should i just restore my notebook and see if it happens again? It's quite confusing and frustrating.

As Linc says, use the phone support, but you may find that Apple tell you a fix is in the pipeline.
There appears to be a glitch with current MBA's where, if it goes to sleep and you try to wake it immediately, it can freeze (search this forum for MBA won't wake from sleep, MBA crash to black screen and similar for loads of info - for example this massive thread https://discussions.apple.com/thread/5118135?start=0&tstart=0). Your description of what happens matches this picture exactly. Until Apple come up with a fix, you can either switch off sleep or resist the temptation to wake the MBA too quickly after it goes into sleep mode.
Some of us (including me) found this unsatisfactory - particularly as Apple are taking a lot longer to find a solution than they originally claimed - and got a refund. In my case I put the money towards a retina MBP.
As for the Chrome issue, I do not and have never used Chrome but was plagued by the problem nevertheless.
By all means get a replacement using the 14 day exchange, but you may well find that the new MBA does the same (this is what I did within the first 14 days and the replacement was no different). If you decide the problem is unacceptable, you could consider a refund - then the choice is yours whether to wait, jump ship to a PC or get a different Mac.

GT70 2PC: 870M does not activate, benchmarks and performance ingame confirm this

I bought myself a GT70 2PC Dominator (the SSD version) with an 870M inside, and after starting several games, checking benchmarks and comparing framerates to my current desktop machine (which has an i5 2500K and a GTX 770), I find that, even after configuring the machine to prefer the "powerful Nvidia processor", and setting all games to use the Nvidia processor, it still uses the 4600 processor. I'm getting rather low framerates, I'm not hearing any fan speed increase, Benchmarks ingame like the one from Sleeping Dogs still report the 4600 GPU and some games won't even boot up properly (like Just Cause 2). I did configure all the games to use the Nvidia GPU in my Nvidia Control panel, I updated my drivers, Device manager is finding both cards.
Is there a way to tell my laptop "Look, I don't care about saving my battery, I don't care about noise, I don't care about heat (as long as it does not damage anything), just ALWAYS use the Nvidia GPU, even when it's not needed, so I can actually have some enjoyment out of the 1600 EUR I spent on this machine"?

Check your power button light and see if it lights red(red=dgpu, white=igpu) and make sure you've attached battery with AC plugged while playing games.
Fan speed is changing by different cpu/gpu temperature, so check the temperature and the fan speed in Dragon Gaming Center and compare the temperature and the fan speed under different situation.
Can you tell the graphics driver version you're using? (both Intel and NVIDIA one)
Sadly for the Optimus structure, you can't disable the single graphics but you should be able to use the dGPU to run the game. Except the settings in NVIDIA Control Panel, you can also try to right click on the shortcut of the game and choose "run with graphics processor"> "High-performance NVIDIA processor"

Poor performance by Matlab and Windows benchmark tests

Hello
I have a Lenovo Thinkpad W520 with Windows 7 64bit installed. Compared to other Notebooks with comparable hardware, my Lenovo is very slow. To prove my thought, I carried through a Matlab (64bit) benchmark test and the Windows 7 benchmark test. In both tests, my Lenovo was worse than the other notebooks. During the tests I set the Power Manager to performance and activated the Lenovo turbo boost.
Now I want to ask, if there are any settings, perhaps in the bios, to speed the laptop up? Or why result such a bad performance although the hardware is very good?
Thanks for your help
Solved!
Go to Solution.

Thank you very much for your reply!
I dated up the BIOS from version 1.27 to 1.32. Now the benchmark tests all look great!!
Thanks

MacBook Pro retina 13" (late 2013, Haswell) why is the highest resolution rendered in 3360x2100, which the system obviously can't handle?!

Hey Apple and Apple users,
I just recently switched from an MacBook Air (mid 2013) to the new Haswell MacBook Pro retina 13" (late 2013 model).
The main reason to do so was the better screen and the option to run higher scaled resolutions, as advertised.
I was willing to trade in the Air's low weight, smaller formfactor and the extra batterylife for the better screen.
When it comes to resolutions:
1280x800 is just a joke nowadays, why is the MacBook Pro still based on it? 13" and 1440x900 work quite fine...see MacBook Air.
1440x900 is okay in everyday life situations, but sometimes you just need "more space" as the third scaling option calls it as well in the display-settings.
1680x1050 on 13" can sometimes be challenging, but it is really fine if you need to work and have a lot UI elements to deal with (Photoshop etc.)
So 1680x1050 scaled on the 2560x1600 screen would be my daily driver for work. So I expected a scaling factor of "1.5238..." (2560/1680) will be used most of the time.
Frankly spoken...I am really disappointed by the 13" retina MacBook Pro and the way OS X Mavericks is handling the scaling.
It obiously works well and responsive with the 1280x800 non-scaling resolution ("best for retina, scaling factor 2: 1280 --> 2560).
Is still "okay-snappy" at 1440x900 (scaling factor 1.777...: 1440 --> 2560).
But it totally breaks down at 1680x1050 (scaling factor 1.523...: 1680 --> 2560).
I use "Mission Control" all the time and it is stuttering and slow...overall a sluggish and unbearable experience.
The same goes for Desktop-Switching by swiping the screens or resizing and moving windows and files around.
So I was wondering why...and then I noticed that when taking a screenshot of my entire screen (CMD + 3) the image size of the screenshot was 3360x2100 and not the expected 2560x1600.
This means Mavericks is not scaling the UI with the appropriate factor (see above). For retina MacBooks it ist simply rendering one pixel of the classic, non-retina Interface in now 4 pixels (2x2) and sending this straight to the screen, which then "scales" it down to the resolution it is able to display.
Why is this bothering me? – Because the machine has to do a lot more work to render the higher resolution that can't even be displayed, due to the screens limitations.
2560x1600 = 4096000 pixels = 100%
3360x2100 = 7056000 pixels = 172%
This means the MacBook has to render 72% more information than actually can be displayed on the built-in screen and are just lost.
Of course this makes the overall UI feel sluggish and idly! Especially with the not that performant Intel Iris IGP and no dedicated graphics card.
All this extra "rendering" just results in a worse battery life due to the extra work that the graphics card and CPU have to do.
First, Apple...are you serious about this?
What can we do about this?
Is there a way to have this properly scaled?
Thanks in advance to everyone.

Okay...I thought deeply about this issue.
Even looking into ways to maybe fixing this on the software side. But it occured to me that this is to deeply embedded into the system that it won't be possbile to come up with a sufficient third-party solution.taht
The problem is that there is no intermediate step inbetween "standard DPI" graphics and "HiDPI" (image-resources named "[email protected]").
But there should be one, to minimize the amount of image information that needs to be processed by the device.
This effects especially the MacBook Pro retina 13", as it has the less powerful Intel Iris IGP, when performing on the highest scaled resolution (equivalent to 1650x1080).
As well as the MacBook Pro retina 15" without a dedicated, second Graphics Unit,when performing on the highest possible scaled resolution (equivalent to 1920x1200).
Those devices don't have enough resources to handle the consequential "HiDPI" resolutions of 3360x2100 for the 13" and 3840x2400 for the 15" model.
(Comment: 3840x2400 = 9,2MP, which is more than the standard 4K resolution of 3840x2160.
I analyzed the scaling factors as follows:
Factor "2" to display 1280x800 on the 13" model and 1440x900 on the 15" model in HiDPI.
Factor "1.777..." would be required to display 1440x900 on the 2560x1600 screen of the 13" model.
Factor "1.714..." would be required to display 1650x1080 on the 2880x1800 screen of the 15" model.
Factor "1.523..." would be required to display 1650x1080 on the 2560x1600 screen of the 13" model.
Factor "1.5" would be required to display 1920x1200 on the 2800x1800 screen of the 15" model.
As all the models seem to have no problem with rendering the intermediate scaling step (1440x900 for the 13" and 1680x1050 for the 15" model) at HiDPI, hence with the scaling factor 2. There is no real need to provide the UI-elements.
But in my opinion there is a strong need for a "MidDPI" ([email protected]) intermediate step for all the UI graphic elements that wis based on the scaling factor 1.5.
This would result in the following:
13" MacBook Pro retina with 2560x1600 screen:
1280x 800 @ 2x = 2560x1600 (no surplus, native screen resolution)
1440x 900 @ 2x = 2880x1800 (the surplus of 1088000px = ~1MP can still be handled and compansated by the weak Iris IGP)
1680x1050 @ 2x = 3360x2100 (the surplus of 2960000px = ~3MP is too much for the weak Iris IGP)
NEW 1680x1050 @ 1.5x = 2520x1575 (with a small border of left=20px, bottom=13, right=20, top=12 pixels, which means ~2,2mm and ~1.3mm around the 1.5x scaled image!).
15" MacBook Pro retina with 2880x1800 screen:
1440x900 @ 2x = 2880x1800 (no surplus, native screen resolution)
1680x1050 @ 2x = 3360x2100 (the surplus of 1872000px = ~2MP can still be handled by the Iris Pro IGP)
1920x1200 @ 2x = 3840x2400 (the surplus of 4032000px = ~4MP can't be handled sufficiently by the Iris Pro IGP)
NEW 1920x1200 @ 1.5x = 2800x1800 (nu surplus, native screen resolution, no border needed!)
The big advantage is that all the graphics have already been remodeled for the HiDPI mode.
So scaling those "factor 2x" images down to "factor 1.5x" should be comparably easy!
Example:
Left: Traditional "standard DPI" icon of "all my files".
Right: "@2x.png" for the "HiDPI" modes.
Middle: "@1.5x.png" suggestions for the "MidDPI" mode; easily scaled down from the lovely HiDPI graphics.
With this simple introduction of the "MidDPI" modes a significant amount of processing power can be saved.
This not only improves battery life when driving those higher scaled resolutions, but also lets the user access those resources for what really matters: processing power for the primary job that the Mac needs to get done.
Thanks in advance.

MacBook Pro Retina i7 2.6GHz quad (Haswell) vs iMac i7 3.5GHz quad for LPX?

Hi guys
I have decided that despite the fact I love my current six-core Mac Pro, it's time to move onto something with a faster internal bus, SSD, and faster RAM… and the new "pot plant" Mac Pro is out of my price range.
I need to have a computer for live work, so I am either looking at getting a Mac Pro Retina 15" 2.6GHz i7 (the highest spec Haswell), or an iMac 3.5GHz i7 (the highest spec iMac) + a more basic MacBook for live use.
Really wondering how the Retina MacBook and the iMac with the above specs would compare, performance-wise. It seems scary to be moving over to either of them from a six-core 3.33GHz processor in the Mac Pro, but I am pretty confident that in a lot of ways, such as latency, either of these will actually perform better than the Mac Pro does.
I would love to hear your thoughts, though, on how the MacBook and the iMac with the above specs would compare.
I use quite a lot of virtual instruments, generally run Logic Pro X (although Pro Tools is also seeing a bit of use), and my projects are pretty large. I use BFD3 and other drum plug-ins with a V-Drums kit, which requires really low-latency performance.
How would these two computers compare in terms of heat generation? Would the MacBook get very hot under extreme workloads?
And in terms of general performance - say, track-count and for single tracks with a software synth (say Diva) and a few effects, played in realtime, how might they compare?
Lastly - will all the heat and tiny fan on the MacBook mean that it is more likely to die after a year or two of use? (One of the things I love about my current Mac Pro is that I can replace any part that dies, quickly and easily - and in any case the processor is cooled by a giant fan!)
So - would really love your thoughts, especially if you are using either or both currently.
Cheers,
Mike

I owned the imac and then returned it for the 2.6 MacBook Pro. After using both there is a noticeable speed difference between the 2. Not a huge difference but there is certainly more lag using the MacBook Pro. I found that lots of the lag went away when I attached to an external display though. At the end of the day I think you just need to decide if you need the portability or not. At first I thought I would keep the imac and then get a cheap MacBook Pro to use on the road but the thought of having multiple Lightroom catalogs sounds very annoying to me. Plus all the transferring back and forth. I will say also that I've been getting a lot of freezes in Photoshop on my mbp and weird errors like the sleep wake failure error some people are talking about when connected to a USB device. I didn't have any of these problems with the imac

VMWare Fusion and Parallels Desktop Benchmark Comparison

This is a quickie benchmark of VMWare Fusion and Parallels Desktop using Super PI, PC Mark 05, and Passmark.
VMWare Fusion 36932
Parallels Desktop 3094 Beta 2
Notes:
Both virtual machines were allocated with large 10+ GB virtual disks and 640MB of RAM. The VMWare CPU was configured with two processors. The Parallels CPU was configured with 1 (two is not available). VMWare reported the CPU as 1 physical, 2 logical processors running at 2.66 GHz while Parallels reported 1 physical, 1 logical processor running at 9.6 GHz (the combined speed of all four cores on the Mac Pro). The max observed CPU utilization in activity monitor when running under VMWare was 200% and max under Parallels was 173%.
I chose not to compare 1 VMWare CPU vs. 1 Parallels CPU. While Parallels does not support SMP or multithreaded processes on multiple processors the CPU utilization on the Mac went well above 1 core (173%). For this comparison, I wanted to see results of max processing based on what the two vendors have delivered, as opposed to benchmarking the underlying "virtual or hypervisor cpu" on a 1:1 basis. This explains why VMWare was 2x faster than Parallels on some CPU tests.
Both of these products are beta. VMWare is running in debug mode (can not be turned off in this beta).
Caveat emptor on these stats. This was an unscientific exercise to satisfy my curiosity. Some of the extraordinary differences are highlighted with <--.
Platform:
Mac Pro 2.66 GHz, 2GB RAM, Nvidia 7300GT
Disk 1 - OS X, 73GB Raptor
Disk 2 - dedicated disk where each virtual machine image was created separate from the OS or any OS-related virtual memory files.
VMWare and Parallels guest OS: Windows XP Professional, SP 2
Comparison Benchmrk
VMWare Fusion 36932 and Parallels Desktop 3094 Beta 2
Super PI Parallels VMWare
512K 8s 9s
1M 20s 21s
4M 1m 57s 2m 03s
PC Mark 05 Parallels VMWare
CPU Test Suite N/A N/A
Memory Test Suite N/A N/A
Graphics Test Suite N/A N/A
HDD Test Suite N/A N/A
HDD - XP Startup 5.0 MB/s 19.54 MB/s <--
Physics and 3D Test failed Test failed
Transparent Windows Test failed 69.99 Windows/s
3D - Pixel Shader Test failed Test failed
Web Page Rendering 3.58 Pages/s 2.34 Pages/s
File Decrypt 71.73 MB/s 67.05 MB/s
Graphics Memory - 64 Lines 179.92 FPS 111.73 FPS
HDD - General Usage 4.82 MB/s 42.01 MB/s <--
Multithread Test 1 / Audio Comp N/A N/A
Multithread Test 1 / Video Encoding Test failed Test failed
Multithread Test 2 / Text Edit 152.85 Pages/s 138.48 Pages/s
Multithread Test 2 / Image DeComp 5.91 MPixels/s 35.4 MPixels/s <--
Multithread Test 3 / File Comp 3.22 MB/s 6.03 MB/s
Multithread Test 3 / File Encrypt 19.0 MB/s 33.26 MB/s <--
Multithread Test 3 / HDD - Virus Scan 27.91 MB/s 25.49 MB/s
Multithread Test 3 / Mem Lat - Rnd 16MB 5.34 MAcc/s 6.63 MAcc/s
File Comp N/A N/A
File DeComp N/A N/A
File Encrypt N/A N/A
File Decrypt N/A N/A
Image DeComp N/A N/A
Audio Comp N/A N/A
Multithread Test 1 / File Comp N/A N/A
Multithread Test 1 / File Encrypt N/A N/A
Multithread Test 2 / File DeComp N/A N/A
Multithread Test 2 / File Decrypt N/A N/A
Multithread Test 2 / Audio DeComp N/A N/A
Multithread Test 2 / Image DeComp N/A N/A
Memory Read - 16 MB N/A N/A
Memory Read - 8 MB N/A N/A
Memory Read - 192 kB N/A N/A
Memory Read - 4 kB N/A N/A
Memory Write - 16 MB N/A N/A
Memory Write - 8 MB N/A N/A
Memory Write - 192 kB N/A N/A
Memory Write - 4 kB N/A N/A
Memory Copy - 16 MB N/A N/A
Memory Copy - 8 MB N/A N/A
Memory Copy - 192 kB N/A N/A
Memory Copy - 4 kB N/A N/A
Memory Lat - Rnd 16 MB N/A N/A
Memory Lat - Rnd 8 MB N/A N/A
Memory Lat - Rnd 192 kB N/A N/A
Memory Lat - Rnd 4 kB N/A N/A
Transparent Windows N/A N/A
Graphics Memory - 64 Lines N/A N/A
Graphics Memory - 128 Lines N/A N/A
WMV Video Playback N/A N/A
3D - Fill Rate Multi Texturing N/A N/A
3D - Polygon Throughput Multiple Lights N/A N/A
3D - Pixel Shader N/A N/A
3D - Vertex Shader N/A N/A
HDD - XP Startup N/A N/A
HDD - Application Loading N/A N/A
HDD - General Usage N/A N/A
HDD - Virus Scan N/A N/A
HDD - File Write N/A N/A
Processor Intel Core 2 9653 MHz Processor Unknown 2661 MHz
Physical / Logical CPUs "1 Physical, 1 Logical" "1 Physical, 2 Logical"
MultiCore 1 Processor Core Multicore 2 Processor Cores
HyperThreading N/A N/A
Graphics Card Generic VGA Generic VGA
Graphics Driver Parallels Video Driver VMWare SVGA II
Co-operative adapters No No
DirectX Version 9.0c 9.0c
System Memory 640 MB 640MB
Motherboard Manufacturer N/A Intel Corporation
Motherboard Model N/A 440BX Desktop Reference Platform
Operating System Microsoft Windows XP Microsoft Windows XP
Passmark Parallels VMWare
CPU - Integer Math (MOPS) 112.35 230.31 <--
CPU - Floating Point Math (MOPS) 280.46 588.33 <--
CPU - Find Prime Numbers (OPS) 446.37 676.99 <--
CPU - SSE/3DNow! (MMPS) 2118.56 4737.13 <--
CPU - Comp (KB/s) 2994.16 5952.34 <--
CPU - Encrypt (MB/s) 18.09 36.27 <--
CPU - Image Rotation (IRPS) 598.21 1184.41 <--
CPU - String Sorting (TPS) 2118.81 3672.59 <--
Graphics 2D - Lines (TPS) 220.71 25.15 <--
Graphics 2D - Rectangles (TPS) 189.74 61.8 <--
Graphics 2D - Shapes (TPS) 39.54 13.71 <--
Graphics 2D - Fonts and Text (OPS) 190.39 75.88 <--
Graphics 2D - GUI (OPS) 439.77 63.72 <--
Memory - Allocate Small Block (MB/s) 2533.83 2526.21
Memory - Read Cached (MB/s) 1960.5 1906.27
Memory - Read Uncached (MB/s) 1871.79 1826.08
Memory - Write (MB/s) 1687.81 1545.43
Memory - Large RAM (OPS) 60.99 46.37
Disk - Sequential Read (MB/s) 102.11 76.45 <--
Disk - Sequential Write (MB/s) 58.33 50.9
Disk - Rnd Seek + RW (MB/s) 51.4 40.4
CPU Mark 711.08 1432.72 <--
2D Graphics Mark 743.31 176.5 <--
Memory Mark 599.94 580.38
Disk Mark 766.11 606.7
PassMark Rating 557.27 637.35 

Thanks for posting these numbers - it's an interesting comparison.
I would expect the final VMWare fusion performance numbers to be quite a bit better than that of Parallels - they have almost a decade's worth of experience more than the Parallels folks in this arena, and a much larger development team to boot.
Once VMWare Fusion is released to the public, I think that you'll see a clearer distinction between the two products. VMWare will continue to appeal to the professional customer, with a more robust feature set and corporate-friendly features (and a correspondingly higher price tag); Parallels will fall more into the consumer/VirtualPC-replacement market. It will be interesting to see how Parallels will be affected when (and if) VMWare player is ported to OS X.
Interesting about the Parallels performance stats on a native partition - looks like almost enough reason to avoid the bootcamp partition approach altogether. Sharing a native windows installation with a VM in parallels is a pretty scary situation in any case, as the two environments have entirely different hardware configurations. Do-able, but there is some black magic involved (if you want to see an example of what I mean, try to move a windows installation from one machine to another w/different hardware sometime - it ain't pretty); I wouldn't try this in a production scheme unless I had REALLY good backups.

Performance benchmark

Hi,
I would like to do some performance benchmarking using Xmath and SystemBuild. Matlab has a tic/toc function, and for Simulink there's model callback.
Is there a similar function in both Xmath and SystemBuild?
Thanks

pkad,
Another thing you might want to take a look at is the Hyperbuild manual. As part of its tutorial it describes a way to time the run-time of a simulation.
Have a great weekend,
NathanT

No keyboard & mouse during Win 7 installation on 13" rMBP (Haswell)

Installing Windows 7 64-bit with Boot Camp isn't possible because the keyboard and mouse aren't working on my new 13" rMBP with Haswell processor. I've downloaded the "Support software" and used the Boot Camp Assistante to create the install USB drive. Adding a USB keyboard and mouse doesn't solve the problem. Has anybody solved the problem or should I wait for Apple to update the "Support software"?

Have seen the update and done what I could with it, again I am stuck on the last step, aka having to enter a name for the partition and choose a color, I have read that i need a wired mouse and keyboard to get past this step. Is that true? I have just been assuming maybe there is something else going on and my system is missing something during that final step. Do I have to suck it up and buy a wired mouse and keyboard?
Another thing.
This download contains the Windows Support Software (Windows Drivers) you need to support 64 bit versions of Windows 7 and Windows 8, and Windows 8.1 on your Mac.
For more information on which operating systems are supported on different Mac systems, click here:http://support.apple.com/kb/HT5634
          • The download file is a .zip file. Double click it to uncompress it, if it is not automatically uncompressed.
          • Double-click the Boot Camp5 folder.
          • Copy the entire contents of the .zip file to the root level of a USB flash drive or hard drive that is formatted with the FAT file system
          • When running Windows, locate the Boot Camp folder on the USB media you created in Step 3 and double click to open it.
          • Double click on setup to start installing the Boot Camp Support Software.
          • When prompted to allow changes, click on Yes and follow the onscreen instructions.
          • Installation can take a few minutes. Don't interrupt the installation process. When installation is complete, click Finish in the dialog that appears.
          • A system restart dialog box appears. Click Yes to complete the installation.
When running Windows, locate the Boot Camp folder on the USB media I created in step three. How would I be able to do this if I cant get past last step of the installation process
Message was edited by: awaddington

Nvidia 180.22: mrxvt transparency and reasonable benchmarks

I just got a 9800GT a few days ago. Just now I upgraded from 173.X nVidia drivers to the newest 180 in the repositories. Two questions:
1) The tabbar transparency in MRXVT has been messed up with the driver upgrade. When I was using 173, it was fine, but using 180 breaks it. Now the tabbar has an odd semi-transparent hue to it, although other mrxvt transparency is fine. IIRC when I tried 177.X I had the same problem. Here's my xorg.conf, do I need to enable anything new with the latest driver? The only transparent-ish thing that has broken is the mrxvt tabbar. I'm not even using real transparency, just pseudo.
Section "ServerLayout"
Identifier "X.org Configured"
Screen 0 "Screen0" 0 0
InputDevice "Mouse0" "CorePointer"
# InputDevice "Keyboard0" "CoreKeyboard"
EndSection
Section "Files"
# RgbPath "/usr/share/X11/rgb"
ModulePath "/usr/lib/xorg/modules"
FontPath "/usr/share/fonts/misc"
FontPath "/usr/share/fonts/100dpi:unscaled"
FontPath "/usr/share/fonts/75dpi:unscaled"
FontPath "/usr/share/fonts/TTF"
FontPath "/usr/share/fonts/Type1"
EndSection
Section "Module"
Load "GLcore"
Load "dbe"
Load "xtrap"
Load "record"
Load "extmod"
Load "dri"
Load "glx"
Load "freetype"
EndSection
#Section "InputDevice"
# Identifier "Keyboard0"
#Driver "kbd"
# Driver "evdev"
#EndSection
Section "InputDevice"
Identifier "Mouse0"
Driver "mouse"
Option "Protocol" "auto"
Option "Device" "/dev/input/mice"
Option "ZAxisMapping" "4 5 6 7"
EndSection
Section "Monitor"
#DisplaySize 340 270 # mm
Identifier "Monitor0"
VendorName "SAM"
ModelName "SyncMaster"
### Comment all HorizSync and VertRefresh values to use DDC:
HorizSync 30.0 - 81.0
VertRefresh 56.0 - 75.0
Option "DPMS"
EndSection
Section "Device"
### Available Driver options are:-
### Values: : integer, <f>: float, <bool>: "True"/"False",
### <string>: "String", <freq>: "<f> Hz/kHz/MHz"
### [arg]: arg optional
#Option "SWcursor" # [<bool>]
#Option "HWcursor" # [<bool>]
#Option "NoAccel" # [<bool>]
#Option "ShadowFB" # [<bool>]
#Option "UseFBDev" # [<bool>]
#Option "Rotate" # [<str>]
#Option "VideoKey" # 
#Option "FlatPanel" # [<bool>]
#Option "FPDither" # [<bool>]
#Option "CrtcNumber" # 
#Option "FPScale" # [<bool>]
#Option "FPTweak" # 
#Option "DualHead" # [<bool>]
Identifier "Card0"
Driver "nvidia"
VendorName "nVidia Corporation"
BoardName "GeForce 9800GT"
BusID "PCI:1:0:0"
EndSection
Section "Screen"
Identifier "Screen0"
Device "Card0"
Monitor "Monitor0"
DefaultDepth 24
Option "NoLogo" "True"
Option "RenderAccel" "True"
Option "AllowGLXWithComposite" "True"
SubSection "Display"
Viewport 0 0
Depth 1
EndSubSection
SubSection "Display"
Viewport 0 0
Depth 4
EndSubSection
SubSection "Display"
Viewport 0 0
Depth 8
EndSubSection
SubSection "Display"
Viewport 0 0
Depth 15
EndSubSection
SubSection "Display"
Viewport 0 0
Depth 16
EndSubSection
SubSection "Display"
Viewport 0 0
Depth 24
EndSubSection
EndSection
Section "ServerFlags"
Option "AutoAddDevices" "False"
EndSection
2) I ran glxgears on the old and new drivers. The old ones yielded about 6800 - 6900 FPS on average. The new ones yield about 6900 - 7000 FPS on average. So, about a ~50FPS improvement, not much to write home about. I've looked through other threads and this seems to be about on par for the graphics card I have (but I hear that on other distros nVidia cards can get much better benchmarks). Just want to check, does that sound about right? I'm unexperienced with tweaking Xorg for graphics performance and I basically just have what nvidia-config spit out.
Thanks.

Pseudo transparency works by placing a copy of your background in the background of your window. (like behind the words in your terminal)
True Transparency will do that, and add an image of your other windows that are behind the transparent window.
If you enable composite, (have a graphics card that can handle it), install xcompmgr and transset-df, then add
xcompmgr -C & # The & keeps it from waiting for the program to stop before doing anything else.
to you autostart, it will enable transparency. I'm using openbox and it works beautifully.
Here is my .Xdefaults (by the way if it doesn't exist, you can create it in your user folder, just don't miss the' . ' !
URxvt*foreground: #A8A8A8
URxvt*scrollBar_right: false
URxvt*scrollBar: true
URxvt*scrollstyle: plain
URxvt*borderLess: false
URxvt*inheritPixmap: true
URxvt*geometry: 124x24
URxvt*tint: white
URxvt*font: xft:monofur:size=8
#URxvt*font: xft:terminus-font:
#URxvt*font: xft:anorexia
URxvt*color0: #000000
URxvt*color1: #A80000
URxvt*color2: #00A800
URxvt*color3: #A85400
URxvt*color4: #0000A8
URxvt*color5: #A800A8
URxvt*color6: #00A8A8
URxvt*color7: #A8A8A8
URxvt*color8: #545054
URxvt*color9: #F85450
URxvt*color10: #50FC50
URxvt*color11: #F2FC50
URxvt*color12: #5054F8
URxvt*color13: #F854F8
URxvt*color14: #50FCF8
URxvt*color15: #F8FCF8
## xterm config
xterm*termName: xterm
xterm*background: #000000
xterm*foreground: #A8A8A8
urxvt*depth: 32
urxvt*background: rgba:0000/0000/0000/dddd #the transparent magic happens here!
URxvt.perl-ext-common: default,matcher
URxvt.urlLauncher: firefox
URxvt.matcher.button: 2
URxvt*matcher.pattern.1: \\b(mailto|http|https|ftp|file):[/]*[\\w-]\\.[\\w./?&@#-]*[\\w/-]
URxvt*matcher.pattern.2: \\bwww\\.[\\w-]\\.[\\w./?&@#-]*[\\w/-
I use urxvt, but if you just change the u to an m, it 'should' work.
Last edited by LeoSolaris (2009-01-29 18:28:15)

Are there any performance benchmark tools for Flash?

I am looking to benchmark Flash on various computers that I use. I was surprised that the performance of Adobe Flash on my Intel i5 computer running Windows 7 Pro 64-bit OS and IE 10 was MUCH WORSE than running on a Windows 7 Pro 32-bit on an Intel i3 computer running the same browser.
I have tried running both 32-bit IE and 64-bit IE and get the same general bad performance on the 64-bit Windows OS. I would like to find a tool to benchmark these various computers so that I can establish baseline performance while I explore finding a fix Adobe Flash on a 64-bit OS.
Can someone suggest some tools for Flash performance benchmarking? Thank you.

The best advise we can really give you is that both companies offer free trials and you should download them both and see which works best for you. I own Parallels Desktop v6, and VMWare Fusion v3. For me, VMWare s better for some things, but Parallels is better for most. Depending on what you do and how you use your applications your milage may vary.
One other note to keep in mind. Since Apple is looking to release a new OS version in the very near future, you might want to hold-off a bit on our vitualization choice just yet. I would exect that both companies will be working on a new release for support/compatibilty of the new MacOS, so you might want to wait to see if there are any other changes that make you want to lean towards one or the other...

After Effects Multi-Core Benchmarks

I have been doing some testing trying to figure out how fast after effects renders and how to
help it render faster. So far i have been very dissapointed with the results. no matter how
much money we spend buying the fastest systems we can i cant seem to get much of a speed
increase. we have 8 computers with 8 cores each now. but i cant seem to get after effects to
use the extra cores even when i have 20Gb ram and enable multi frames with 2GB per frame. i see
it load all the extra copies in task manager but when i render each time 1 core has "some"
usage and the other 7 are always around 10-15% usage.
so i wanted to try a simple benchmark that everyone could try and post their results.
so i made a ntsc dv composition default at 30 seconds and just render it. NOTHING, just blank
frames of nothing. how fast can afx output data like this? i tried tests with multiple frames
enabled and disabled and output to tiff files (no compression) or the microsoft DV 48khz
preset, both with the default BEST setting.
Now i understand that after effects and premiere have 2 completely different rendering methods
but still it is worth pointing out that premiere will output 30 seconds of blank video or
actual real dv video footage to a DV AVI file in about 3-4 seconds. so why is it the same
machine takes 10 times longer to render from after effects?
I know in premiere i can simple drop in a dv avi file and export to mpeg2 and i can watch all 8
cores almost max out as it renders about 6X faster then realtime.
How can i do something in after effects to see my 8 cores max out?
Please give any tips or tricks to speed up after effects. We must use vista64 as we have a 30TB fibrechannel array.
Dell Laptop M6300 - Core 2 Extreme x9000 @2.8ghz (2 cores)
Adobe CS4 Windows XP 64 bit - 8GB ram
Multiple OFF     Tiff=1:24
                       DV=1:24
Multiple ON      Tiff=1:32
                       DV=1:30
Dell Precision 690 - Dual Quad Core Xeon E5320 @1.86ghz (8 cores)
Adobe CS4 - Windows Vista 64 bit - 4GB ram - Matrox Axio LE
Multiple OFF     Tiff= :47
                       DV= :43
Multiple ON      Tiff= :56
                       DV= :52
Dell Precision T7400 - Dual Quad Core Xeon X5482 @3.2ghz (8 cores)
Adobe CS3 - Windows XP 32 bit - 4GB ram - Matrox Axio LE
Multiple OFF     Tiff= :30
                        DV= :30
Multiple ON      Tiff= :31
                       DV= :30
Dell Precision T7400 - Dual Quad Core Xeon X5482 @3.2ghz (8 cores)
Adobe CS4 - Windows Vista 64 bit - 20GB ram - Matrox Axio LE
Multiple OFF     Tiff= :30
                       DV= :31
Multiple ON      Tiff= :35
                      DV= :35

Well we can toss around reasons for AE not using a processors full potental on a comp, but all I know is that all of the truly multithreaded and multi-processor enable applications I use are much better at using resources to their fullest than AE, or for that mater, most of the programs in the MC.
When I run those programs my system is pushed to the limit- which is why I bought a quad core system in the first place. Mental ray, Fusion, 3D Coat, Zbrush...the list is long of programs that have no problem using all my cores for 90%-100% of opperations.
In the end it just adds up to the fact that Adobe owns a large corner of the market- and since there is no competition, sees no reason NOT to be 5-10 years behind the curve when it comes to resource managment in their software.
Making maters worse is how a lot of the user base is oblivious to the technological changes in processors over the last five years. These people don't know that all but one of their cores sit idle most of the time, and they buy the corp. speak put out by Adobe about "...how complex every thing is- so you don't understand...". Sorry- I may not be a programer or a processor engineer for Intel or AMD, but I know when a program is using resources or not and I know quite a few of the things Adobe has said are "...just too complicated to do..." are really covers for lack luster R and D. Either your programers need to get up to speed, or Adobe needs to actually do the right thing and set more money aside for development. I'm betting it's the later.
Softimage 7.x is fully multithreaded and 64bit (yes all the way through not just with mr). This is a complicated program- and the development team is probably 1/10th the size of that working on PS. So why after all of these years are we still waiting for even a half baked attempt at such things on the Adobe front?
The way AE handles RAM compared to programs like Fusion and the like is pathetic.
Don't get me wrong- I love the program for motion graphics and simple comp work, but again, the resource management with AE feels like I'm back in OS8.
-Gideon

Haswell E benchmarks

Similar Messages

Maybe you are looking for