Hello,
I’m looking for a new PC to work on 3dMax and Unreal Engine for Architectural visualization. I need power to build the light. Now i have a question about the CPU, i’m not sure to buy a intel or new AMD.
I moved from 4 core intel to 8 core amd (1700), single core speed like same but material compilation and lighting build is 2,5x faster.
so I recommend you to wait for amd threadripper - 16 core 32 threads processor, it comming soon.
hmm, i though if you are getting 1080ti you are not on budget but i think that card is too powerful for what you want, its + - for 4K gaming, so IMO get 1070 and rest of money put into CPU
2.5x faster lightmap building? The benchmarking from the other Ryzen thread showed the 8 core AMD processors are basically the same speed as the 6 core i7s at lightmass baking. If lightmass scaled perfectly with those numbers, it’d be only 1.5x faster.
Well im probably not very accurate, but my old PC (intel 750 - 3GHz, from 2009) made material shader in 3-5s, but with ryzen its done in 1-2s.
Lightmass (that lowest / fastest option) - i compared my PC with one tutorial where guy opened project 2x faster and compiled code 2x faster (i think he have some kind of fast 4-core intel) but when it came to build lighting, my PC was faster around 2 - 2,5x. it took him maybe 1 minute but i was done it in 20-30s.
My comparison of CPUs is not good at all but im really satysfied with 8 core ryzen, i feel that speed when it comes to multi-core usage…
edit: im running ryzen only on 3 GHz and RAMs on 2133 MHz (3k mhz rams), i plan to overclock CPU and RAM too but im OC noob so i didnt found good tutorial yet how to OC it properly
While I can vouch for the R7 1700 being an extreme amount faster than the i5 750 as I also had this upgrade path, the 750 is not exactly the kind of CPU we’re talking about here =p
Stock 1800X should be faster than stock 6800K.
Honestly I find it hard to recommend one over the other, if you do OC, the 6800K is probably the better buy but at the same time the 1700 is a much better buy than the 1800X, again, IF overclocked.
If you don’t want to OC, and just want something fast for building lightmass & shader compilation, and don’t want to pay for the 8+ core Intel CPUs, buy the 1800X.
In that benchmarking thread I’m only running the 1700 at ~3.5Ghz for most of the tests, while all other CPUs were OC’d to over 4Ghz.
I didn’t know this before, but apparently visual studios linker is single threaded, which explains my CPUs surprisingly poor code compilation results, by overclocking to 3.5Ghz, I actually lost 200Mhz of single core speed, so a higher clocked Ryzen will be pretty competitive there as well.
It’s all too hard, sorry for not really helping =p
If there’s a Microcenter near by, it’s a $30 price difference between the i7 6800K and the Ryzen 1700. The 6 core 6800K Intel and the 8 core 1700 perform roughly the same in multithreaded benchmarks, but the i7 performance out performs in single threaded tasks. Those benchmarks do reflect in real world performance with compiling and baking in UE4 from that Ryzen thread.
The 3D max works with multiples of 4 cores, so the Intel 6800 will only use 4 cores instead of 6. So AMD will have twice as many functional cores, in this case being unbeatable.
Yes, UE4 will use as many cores as it can, as long as there’s enough tasks to split. Sometimes at the end of bakes, it’s just a few cores left working.
Reviews so far have been good for the TR. Gaming, meh…But, who purchases a 1k CPU for gaming. Content creation seem to be its high-point. Think I want to get one!
This review is really interesting.
To bad that AMD loses so hard against the Intel. I would have been happy to save a little bit of money.
But now it is impossible for me to buy a Ryzen CPU.