Holomark 2 Released!

This is for a used PC I just bought, not bad for $370

Holomark 2 v2,2,03

Total Score: 23734
Total Runtime: 148.4 sec

GPU scores: 13400
GPU_01 - 506.30 fps - Cube 4 tests
GPU_02 - 30.20 fps - UDT Shape
GPU_03 - 58.80 fps - Wireframe
GPU_04 - 35.30 fps - Shaded
GPU_05 - 26.70 fps - Rendered
GPU_06 - 26.00 fps - Block Rendered
GPU_07 - 13 units Nurbs @ 5 fps in Wireframe
GPU_08 - 8 units Nurbs @ 5 fps in Shaded
GPU_09 - 9 units Nurbs @ 5 fps in RenderSpeed
GPU_10 - 91.70 fps - RenderMesh Render
GPU_11 - 40.80 fps - RenderMesh RenderSpeed
GPU_12 - 57.10 fps - JoinedMesh Render
GPU_13 - 31.90 fps - JoinedMesh RenderSpeed
GPU_14 - 2 units mesh @ 15 fps in Shaded
GPU_15 - 4 units mesh @ 15 fps in Render
GPU_16 - 3 units mesh @ 15 fps in RenderSpeed
GPU_17 - 37.50 fps - mesh in Rendered Studio
GPU_18 - 8.80 fps - Nurbs in Rendered Studio
GPU_19 - 21.10 fps - Block Illustration
GPU_20 - 43.50 fps - 2D single
GPU_21 - 6.40 fps - 2D massive (20x)

CPU scores: 10334
CPU_01 - 13.47 sec - Booleans and Contours
CPU_02 - 1.98 sec - Twist and Taper (UDT)
CPU_03 - 8.15 sec - Meshing Mini
CPU_04 - 0.05 sec - Extract Render Mesh
CPU_05 - 0.09 sec - Join Render Mesh
CPU_06 - 17.44 sec - Reduce Mesh
CPU_07 - 2.43 sec - Calculating Technical display
CPU_08 - 4.77 sec - Making Silhouettes

Gigabyte Technology Co., Ltd.
GA-78LMT-USB3

NVIDIA GeForce GT 640 - 2048.0 MB
DriverVersion: 10.18.13.6510

AMD FX™-6300 Six-Core Processor
NumberOfCores: 3 NumberOfLogicalProcessors: 6
MaxClockSpeed: 3.5 GHz

TotalPhysicalMemory: 16.0 GB

Microsoft Windows 10 Home

  • None - 64-bit
1 Like

MSI’s GS60 - a 15" laptop.
AA off - but clocked at 3178 with AA on.

Holomark 2 v2,2,03

Total Score: 37889
Total Runtime: 209.76 sec

GPU scores: 25650
GPU_01 - 597.00 fps - Cube 4 tests
GPU_02 - 36.60 fps - UDT Shape
GPU_03 - 67.60 fps - Wireframe
GPU_04 - 41.80 fps - Shaded
GPU_05 - 30.80 fps - Rendered
GPU_06 - 27.50 fps - Block Rendered
GPU_07 - 15 units Nurbs @ 5 fps in Wireframe
GPU_08 - 10 units Nurbs @ 5 fps in Shaded
GPU_09 - 32 units Nurbs @ 5 fps in RenderSpeed
GPU_10 - 102.00 fps - RenderMesh Render
GPU_11 - 169.50 fps - RenderMesh RenderSpeed
GPU_12 - 129.90 fps - JoinedMesh Render
GPU_13 - 181.80 fps - JoinedMesh RenderSpeed
GPU_14 - 6 units mesh @ 15 fps in Shaded
GPU_15 - 6 units mesh @ 15 fps in Render
GPU_16 - 202 units mesh @ 15 fps in RenderSpeed
GPU_17 - 83.30 fps - mesh in Rendered Studio
GPU_18 - 10.40 fps - Nurbs in Rendered Studio
GPU_19 - 22.40 fps - Block Illustration
GPU_20 - 59.20 fps - 2D single
GPU_21 - 7.80 fps - 2D massive (20x)

CPU scores: 12239
CPU_01 - 12.23 sec - Booleans and Contours
CPU_02 - 1.87 sec - Twist and Taper (UDT)
CPU_03 - 5.27 sec - Meshing Mini
CPU_04 - 0.03 sec - Extract Render Mesh
CPU_05 - 0.06 sec - Join Render Mesh
CPU_06 - 14.96 sec - Reduce Mesh
CPU_07 - 2.18 sec - Calculating Technical display
CPU_08 - 4.25 sec - Making Silhouettes

Micro-Star International Co., Ltd.
GS60 6QE

Intel® HD Graphics 530 - 1024.0 MB
DriverVersion: 20.19.15.4331
NVIDIA GeForce GTX 970M - 3072.0 MB
DriverVersion: 10.18.13.5374

Intel® Core™ i7-6700HQ CPU @ 2.60GHz
NumberOfCores: 4 NumberOfLogicalProcessors: 8
MaxClockSpeed: 2.6 GHz

TotalPhysicalMemory: 16.0 GB

Microsoft Windows 10 Home

  • None - 64-bit

Rhino 5 sr 12 64 bit

Tried my new dell

After disabling switchable graphics rhino seems a lot faster…

Asus Workstation. X4 antialias

Hello everybody.

Here’s my current laptop configuration:

We’re planning on buying a new powerful desktop machine. How much better would it be to go with Xeon CPU and Quadro GPU instead of top of the line i7 and GeForce 980? Significantly better or just slightly better?

Here are the 1070 results on my “older” dual Xeon system:

No AA:

Notice how the mesh scores drops when I turn on AA:

2xAA:

4xAA:

8xAA:

So as you can see, unlike with Quadros, those pure mesh scores are strongly limited by turning on AA.
This doesn’t affect the normal Rhino nurbs modelling, but if you work with heavy architectural models with lots of meshes, then you will notice the difference.

School’s out! Swapped out the K5000 for a Titan X Pascal, everything else is the same. I posted a comparison. I leave AA at 4x.

So having a quadro card or firepro you say that working with meshes is much better than GTX or AMD counterpart?

Yes but it only applies to nVidia and if you have AA turned on (Which is the default).
AMD doesn’t cripple their cards in the same way, but their AA isn’t very nice and they have a long history of bad OpenGL drivers that can cause various issues and display artifacts.

For a stock, base-model i3 6100 CPU I’m pleased with the score. Most articles on the web indicate this CPU is easy to overclock by 15-20%. I’ll post test results from that once I’ve got it all sorted out and stable. The OS hard drive on this PC is a Samsung nvme 256GB plugged straight into an M2 socket on the underside of the motherboard. Early impressions are that it’s fast, but it can get a bit hot. Total system power draw at the wall socket, including 1x Dell 3007 30” monitor + 2x Dell 2007FP 20” monitors + 2 further SSD’s never goes over 260W. If I had LED monitors, that figure would be closer to 150W. Not bad for a 4960 x 1600 desktop

Hi there! I’ve been using Holomark for about a year now and it’s been a great way to benchmark different system configurations. Well I just built a new computer and at first Holomark was working fine for me (like it always has in the past) but after changing some settings on my computer, including a modest overclocking of the cpu and setting my Quardo global preset to ‘Workstation App - Advanced Streaming’, Holomark now never finishes and always gets hung up at the exact same point (please see screenshot). My computer doesn’t freeze, crash or anything like that. It’s simply that holomark strops running. Even when I go back to my original PC settings with no overclock and setting my nvidia settings back to the default ‘base profile’ preset I encounter this same problem. I even tried installing an older version of Rhino 5 (SR9) and holomark continues to stop working at the exact same point. Does anyone know what might be causing this issue?

Thanks!

And if it helps, here was one of my earlier, successful attempts at running Holomark on this new computer:

Update: Well I continue to have the same issue where Holomark doesn’t finish but maybe 1 in 10 tries it will actually complete and give me a score. My scores have varied pretty significantly, even with the same GPU and OC settings, ranging from about 75,000 to 100,000. Below is the highest score I have received yet with all default settings in the nvidia control panel and a 25% overclock (4.5 GHz) on my CPU. I’d like to do some testing with different nvidia settings but given the troubles I’m having with Holomark right now it’s unfortunately too time consuming.

So we have here the followings:

  1. ASUS Z97-m PLUS (SUPERB OC FOR XEON)
  2. Xeon E3-1231v3 OC @4GHz on 8 cores (doesn’t show correctly here on Hollomark stamp)
  3. 32GB KINGSTON BEAST XMP 1600MHz
  4. Asus STRIX GTX 750ti 2GB @1404 Mhz OC

Just changed the video card.

Nice!
The image is nicely antialiased, did you get that score with aa on?
And if so did you use any special settings?

Hello,

Any idea why holomark score dropped this much after win10 “anniversary” update?

Before:

After:

Thanks!
Janne

Edit: Just saw that the driver version is differend, can that be the cause…? If so, can someone tell me how to roll back that driver. Thanks again!

I have no idea, I see almost all of your CPU scores are lower, but GPU score 14 and 15 are higher. Maybe a background process is running. It wouldn’t be the first time an upgrade requires some background processing to complete the changes it has made.

Unfortunetly aa was set to 0 to get this score. :confused:

The 1070 is surprisingly much lower than 980 which scored 58000 in the following post.
The brand new Titan X Pascal only got 42000 in one of the previous post.
What’s happening with pascal based cards?

Why is the CPU score of 6700HQ so horrible :< It’s even much lower than i7-6650U’s 13340 on Surface 4 Pro. Something wrong with UDT.
Tested 6700HQ on a All-in-one at uni, CPU benchmark seems more reasonable.