Motherboard Daz recommendation

Super cheese sandwich

There is a big difference between using the last chipset x99 and others? Almost all motherboards have 16 lanes per gpu. I ask this because in my country all things cost 4x more than usa. If a titan cost 1k then in my sh*t country is 4k  Same with motherboards and technology things...

 

"""""- Make sure your motherboard has the latest Intel chipset (currently the X99)

      - For graphics applications 32GB of RAM is a good minimum, most high end systems have 64GB but this will have little to no impact on your render time.

      - 6 core CPUs are enough if you are going to use Iray (or other GPU renderer), when you get into high core count GPUs, the CPU becomes less important.

      - Get as many large GPUs as you can… 4GB or memory is a minimum, over 8GB is probably not going to be needed. Core count is critical, so if two less expensive cards give you more cores than one expensive card, go for the two cards…

·        Make sure your motherboard offers at least 8 lanes of PCI per GPU installed (Ideally 16 lanes per GPU). If you have 16 lanes per GPU you get the full performance, having only 8 lanes per GPU gets you around 80% of the GPU performance, this influences the decision to buy one large card over multiple smaller cards

https://helpdaz.zendesk.com/hc/en-us/articles/207530513-System-Recommendations-for-DAZ-Studio-4-

 

Thanks

StratDragon

these are suggestions, not requirements. Yes, the latest and greatest parts will make your 3D experience as thrilling as a unicorn that farts rainbows but there are plenty of users here who's systems do not approach these qualifications

For graphics a 32/64GB system would be great, but the majority of users here are working with far less. 8 or 16GB RAM will allow you to accomplish some amazingly involved renders. The more complex your renders get the more RAM you may wish to have, or if you do a lot of multitasking it's nice to have it available, but its it a necessity? No.

I use a 4 core CPU, it's 7 years old, it gets the job done. If I begin a render before I go to bed it's either close to done or done before I get up the next morning and I tend to render 2400x3200 so your statement is again more suggestion but not a show stopper.

4GB GPU RAM is suggested by the developers if you wish to use CUDA in GPU assisted renders. It's not required. If two cards give you more cores will your PSU give you power enough for the two cards or do you need to by a new PSU along with them. There are exceptions to this rule not limited to the PSU.

Again, if you plan to use GPU assisted renders a MoBo that takes advantage of the latest architecture will be more of a benefit, but the "80%" is an estimate. A PICe3.1 card in a PCIe2.0 slot will work at 2.0 speed. If a 20 minute GPU render takes me 23.5 minutes because I saved a few hundred on a lesser mobo to render it's not any reason to throw my computer out into the street, and it wont stop me from being able to use the computer for 3D work.

You don't even need an Nvida GPU to use Iray, but if you have money to spend on one and would like to render faster it's an option, but hardly a requirement. 

Super cheese sandwich

Thanks for answer StratDragon! 

Super cheese sandwich

It's me again. I was reading about motherboards with more slots (all x16) and I found that one card works with x16 but If I add three cards then I'll get x8 and even x4... Will I feel the difference? Thanks!

Any good motherboard with three slots?

hphoenix

PCI-E lanes (the x16, x8, x4, x1) are a matter of CPU and Northbridge chip.  Each can provide a number of PCI-E lanes to be used by cards/peripherals.

Most CPUs and MBs only provide 24-32 lanes.  Since ALL PCI-E cards and peripherals have to share these lanes, this means that only a limited number can be used by any one device.  If you have 24 lanes available, and you have one PCI-E x16 card, a PCI-E x4 SATA controller (usually on the Motherboard, internally), and you plug in a second x16 card, you no longer have enough lanes available.  So the cards drop to x8 each, and you are still able to run them both, though with lowered bandwidth to each.

That means moving data onto and off of the cards will be slower.  Usually, this means loading time increases (for games) and setup time increases (for renders).  The difference is mostly imperceptible, though when shuttling lots of small data back and forth (gaming) it can have a more noticable impact.

The Motherboard needs to support the addtional lanes to each slot, but more importantly, the CPU needs to have enough lanes available to use them.  The high-end XEON and i7 CPUs can have 32 or more lanes right on the cpu (this usually means a 2011-v3 motherboard, LGA 1151/1150 CPUs don't have enough pins to support the additional lanes) but lower CPUs won't have that many.

So the ASUS Pro 2011-v3 motherboards, paired with a good XEON or high-end i7 cpu will give you the ability to run 2 or 3 x16 cards at full bandwidth.  It'll also probably cost you well over $1000 just for the CPU/Motherboard.

 

Super cheese sandwich

Thank you very much for the detailed answer hp! I'll be very greatful If you can give me a tip for my pc update please!. Which pack it's the best for fast rendering?

 

Pack 1:

Intel 4790

motherboard with 3 slots for graphic cards

2x 980 ti

1x 970

---------------------------------

Pack 2

Intel 6700k

Motherboard  ASUS Pro 2011-v3 with 4 slots for graphic cards

1x 980 ti

1x 970

 

Many many thanks!!!