![]() If you don’t have enough it’s going to slow you down a lot because, you computer will have to swap files in and out of ram from your hard drive. They will just sit there with nothing to do. But once you have enough, adding more won’t help. If you don’t have enough of them you REALLY need more, otherwise everything grinds to a halt. Think of ram like waiters at a restaurant. This is true to some extent, but its not like adding more ram will make your existing system faster. check this stuff yourself.Īlso i notice some people think more ram will make your system faster. If you are getting an AMD ryzen or threadripper, google ” your processor name + what is the best ram” there are some speed matching issues that crop up with the AMD architecture. Check what your processor supports without overclocking. But also don’t go overboard and buy the fastest, because without overclocking you’re not going to feel the difference. Careful don’t cheap out and get the slowest ram, that noticeably lags your workflow. Just get 16GB for now, but make sure your motherboard supports at least 64GB. For OS and other stuff use the SATA drives. So put stuff that needs to be fast on them. Most motherboards support 2 or 3 of them. And M.2 ssd’s which slot in to your motherboard. There are 2 main types to concern yourself with SATA ssd’s which plug in with a cable, most motherboards support about 6 of them. Adata are ussually a good budget choice, otherwise go Samsung i think. Don’t stress too much on which one, just check tom’s hardware for recommendations, and get that one. I wrote a more expansive article on what you need to know for GPU rendering here: ///gpu Hard Drives ![]() so you have to check that on your CPU spec sheets. Some of those PCIe lanes are reserved for nvme drives and other stuff. It is recommended you have 8x for each card, but apparantly some engines allows you to run them at 4x lanes each. ![]() The amount of lanes your CPU has affects the amount of GPU’s you can run. But honestly its rare that scenes are that large. Or in the cases where you can use it, is slows down significantly. When rendering with a GPU renderer you want as much ram as possible, because you cant use your system ram. So i’d buy a motherboard with at least 3 pci-x slots. I think rendering is moving mostly towards the GPU. ( maya apparently benefits from Quadro cards). Quadro cards have no real benefit in Cinema4D or the GPU renderers. I’ll be writing an article on them later. Not because its faster, but because all these awesome GPU renders that are out run on CUDA, and thats an Nvidia technology. So a 5 year old CPU at 3.2 ghz won’t necessarily match a 3.2ghz cpu from this year. Thats because its a newer generation and its more efficient. If you check cb scores, you might notice that a 6700k scores more than a 4770k even though it has a lower clock speed. There is also a cinebench R20 now, but there aren’t many cpu reviewed on it yet, so the scores aren’t as useful.Īnd Just a final note to make this all more confusing. Also make sure when googling for cpu scored, that you are looking at cinebench R15 results, not R11.5 ones. The xeons usually score significantly less on those.Ĭheck // for how CPU’s score in both single and multi core. check the single core benchmarks and compare. But the previous issue with clock speeds is still in play. They only really make sense if you are going the Dual cpu route. For example the viewport in cinema4d, Xpresso, character rigs, and most of the realtime stuff you see runs on one single CPU core, so you will get a more responsive viewport while working with a 6700K, but your renders will go slower ( assuming you are rendering with a CPU based renderer, more on that next). And unfortunately lots of stuff still isn’t multi threaded and runs on only one core. But that is only true if what you need the power for is multi-threaded. Generally it seems to make sense to go for 6x 3.4ghz, rather than 4x 4ghz. For example the i7 5820K wich has 6 cores, but runs at 3.4ghz and then you have the i7 6700k which has only 4 cores but runs at 4ghz. You have processors with lots of cores, but lower core clock speed. So there are 2 main factors to consider when picking a CPU. But for simulating particles and dynamics, still very useful. Also Since GPU renderers are becoming more and more capable, signficanlty less rendering is being done on them. They have great single and multi core speeds. Update 2019: Amd ryzen and Thread rippers have completely dethroned intel performance wise. And more often than not either building their own workstations or at least picking the components and getting someone to build one for them.Ī couple of questions keep coming up in the forums, so i thought i’d write a short article about the main questions. With the inevitable abandonment of the professional market by Apple lots of people are switching to windows.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |