good or bad idea: 5.5K euro for machine 2xTitanZ

Generic forum to discuss Octane Render, post ideas and suggest improvements.
Forum rules
Please add your OS and Hardware Configuration in your signature, it makes it easier for us to help you analyze problems. Example: Win 7 64 | Geforce GTX680 | i7 3770 | 16GB
User avatar
Laticis
Licensed Customer
Posts: 154
Joined: Sun May 02, 2010 2:23 pm
Location: Darwin, Northern Territory, Australia
Contact:

Okay...new setup.

Revo SSD on PCI-E (1) (x16) - big waste of 8 data lanes but that is a small thing for now.
Have discussed back installing this drive with ribbon through the PCI-E (5) which will free me up the PCI-E (1)(x16)
This could be a promising aspect potential third card :D

Titan Z on PCI-E (2) (x8)

Titan Z on PCI-E (4) (x8)

New test results - as expected:
2014-11-15_2136.png
For what ever reason this set-up worked. Windows and Octane have both registered the change and accepted this config.

Next step now the panic is over - get the NAS working so I can really give the system a run for its money.

Would like to say thank you very much for your assistance and feedback, much appreciated.

Cheers Ady
User avatar
Seekerfinder
Licensed Customer
Posts: 1600
Joined: Tue Jan 04, 2011 11:34 am

abstrax wrote:...since we minimized the amount of data that needs to be transferred via the PCI-E bus.
Hi Abstrax,
Is this since v2.0? I probably missed the release notes but could you please expand on that? The full scene obviously has to load. What kind of additional i/o activity is there usually? Which of that did you change and is the difference significant?

We're researching expanders and this info would help.

Thanks,
Seeker
Win 8(64) | P9X79-E WS | i7-3930K | 32GB | GTX Titan & GTX 780Ti | SketchUP | Revit | Beta tester for Revit & Sketchup plugins for Octane
User avatar
abstrax
OctaneRender Team
Posts: 5510
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Seekerfinder wrote:
abstrax wrote:...since we minimized the amount of data that needs to be transferred via the PCI-E bus.
Hi Abstrax,
Is this since v2.0? I probably missed the release notes but could you please expand on that? The full scene obviously has to load. What kind of additional i/o activity is there usually? Which of that did you change and is the difference significant?

We're researching expanders and this info would help.

Thanks,
Seeker
Nothing major changed since the big refactoring of 1.0 beta 2.47 (from 2011). What we do is upload the scene data (which is NOT of the same size as the source data) and then only update that data that changed. During rendering we pretty much transfer only result images from the GPU.
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
User avatar
Seekerfinder
Licensed Customer
Posts: 1600
Joined: Tue Jan 04, 2011 11:34 am

abstrax wrote:
Seekerfinder wrote:
abstrax wrote:...since we minimized the amount of data that needs to be transferred via the PCI-E bus.
Hi Abstrax,
Is this since v2.0? I probably missed the release notes but could you please expand on that? The full scene obviously has to load. What kind of additional i/o activity is there usually? Which of that did you change and is the difference significant?

We're researching expanders and this info would help.

Thanks,
Seeker
Nothing major changed since the big refactoring of 1.0 beta 2.47 (from 2011). What we do is upload the scene data (which is NOT of the same size as the source data) and then only update that data that changed. During rendering we pretty much transfer only result images from the GPU.
Abstrax,
Thanks for clarifying. Nothing new then.

As a matter of interest, what is the typical difference between the source data and the data uploaded to the GPU memory?

Cheers,
Seeker
Win 8(64) | P9X79-E WS | i7-3930K | 32GB | GTX Titan & GTX 780Ti | SketchUP | Revit | Beta tester for Revit & Sketchup plugins for Octane
User avatar
rappet
Licensed Customer
Posts: 1962
Joined: Fri Apr 06, 2012 3:57 pm
Location: The Netherlands
Contact:

rappet wrote:Hi guys,

Help... :roll:
Please, try and talk this out of my head and give me good reasons (other configs) not to buy a machine with 2x TitanZ when the :twisted: in my head wants to spend 5.5K euro.
(probably it will be 6.1k with upgrade RAM and CPU water,
but basically it is 5.5K with i7-5820K, 16GB, 2TB, 250GB SSD
)
I have seen other threads about TianZ, 780's adn more,
but I did not want to hack someone's thread.


greetz,
OK, after almost two years hereby a post ..sort of closing down relatively tothe first post :lol:
You guys did not talk ithe idea out of my head... Due to other build (slave machine with double TitanZ), this build ended up with the TitanZ and a Black and X as upstaries neighbours :) Nicely with open space between the cards.
To pleasure the cards I recently gave them a view outside by replacing the side panel with a custom made panel by Mnpctech.
So, this is the final look...

11 months ago I received the panel, but only recetly I took 15 minutes to swapp the standard side panel for the Mnpctechwindow panel for the Cosmos II Case.
The panels are machined in Cast Acrylic with edges polished by hand.
It was peanuts swapping the panel and installing the 2 Mnpctech fan rings.
The next thing to do is design a 120mm logo to print 3D and install it together with a dust filter on the inside behind the fan rings... after finding another 15 minutes or more. And remove the standard fan cable that you see and add LED strip inside.

Resume: it was a very good idea!

cheers,
Attachments
Tapperworks-Mnpctech-Modding.JPG

4090+3089ti & Quad 1080ti
ArchiCAD25, ofcourse Octane & OR-ArchiCAD plugin (love it)
http://www.tapperworks.com
http://www.facebook.com/pages/TAPPERWOR ... 9851341126
http://www.youtube.com/user/Tapperworks/videos
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Simply beautiful, elegant! Great job. And the panel looks great.
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

PS. What are your core clocks on Z?
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
User avatar
rappet
Licensed Customer
Posts: 1962
Joined: Fri Apr 06, 2012 3:57 pm
Location: The Netherlands
Contact:

smicha wrote:Simply beautiful, elegant! Great job. And the panel looks great.
Thanx.
smicha wrote:PS. What are your core clocks on Z?
1045/1058 mhz (no OC)

Edited:
How does this work?
According to the specs of the Titan Z on nvidia.com I see the Boost Clock 876 mHz, but on Precision X I can see GPU Clock running at 1054 mHz?

Cheers,
Attachments
TitanZ-Clockspeed.JPG

4090+3089ti & Quad 1080ti
ArchiCAD25, ofcourse Octane & OR-ArchiCAD plugin (love it)
http://www.tapperworks.com
http://www.facebook.com/pages/TAPPERWOR ... 9851341126
http://www.youtube.com/user/Tapperworks/videos
Post Reply

Return to “General Discussion”