+- hashcat Forum (https://hashcat.net/forum)
+-- Forum: Misc (https://hashcat.net/forum/forum-15.html)
+--- Forum: Hardware (https://hashcat.net/forum/forum-13.html)
+--- Thread: Ryzen 1800X CPU with 1080 Ti GPU (/thread-6534.html)
Ryzen 1800X CPU with 1080 Ti GPU - N0vajay05 - 05-05-2017
I've read lots on these forums that you don't want to use AMD video cards with Hashcat, so I'm thinking of going with a single 1080Ti FE to start and going with dual in the near future. For the CPU however, I'm leaning towards an 8 core Ryzen to go with some video and photo processing programs I also use that take advantage of the 8 cores. Is there any reason to avoid this and stick with an Intel CPU for hashcat reasons if the processing is all taking place on the GPU?
RE: Ryzen 1800X CPU with 1080 Ti GPU - epixoip - 05-06-2017
1080 Ti is an excellent choice. The best choice, in fact.
If you plan to use the CPU for cracking then I'd avoid Ryzen. Its AVX2 is a lie and they dropped XOP.
RE: Ryzen 1800X CPU with 1080 Ti GPU - NikosD - 06-04-2017
Ryzen 7 1800X is the best mainstream CPU for everything, even cracking.
RE: Ryzen 1800X CPU with 1080 Ti GPU - royce - 06-05-2017
If you're using your CPU for activities indirectly related to the actual cracking phase (like working with wordlists, etc.), then most CPUs are fine (faster and more cores are better, etc).
But if you want to use the CPU for actual cracking, Ryzen is a step backward, just as epixoip stated.
RE: Ryzen 1800X CPU with 1080 Ti GPU - epixoip - 06-05-2017
@NikosD - the 1800X is absolutely not the best mainstream CPU for password cracking. Not even a little bit. We've been over this.
RE: Ryzen 1800X CPU with 1080 Ti GPU - NikosD - 06-05-2017
I tried to find figures and not opinions regarding cracking performance of Core i7 7700K and RyZen 7 1800X using hashcat in this forum, but didn't find any.
After a lot of optimizations regarding memory support of RyZen which plays a significant role even in CPU performance due to CCX-to-CCX communication, maybe it's time for someone to publish some updated benchmarks of RyZen 7 vs Core i7 with a faster DDR4-3200 RAM
Of course, the question of @N0vajay05 has nothing to do with that, because he will run hashcat on the GPU which is absolutely sane and the CPU has no role at all on this.
But because epixoip and others insist on RyZen 7 being slower in CPU cracking than Core i7, show us some figures to be sure.
RE: Ryzen 1800X CPU with 1080 Ti GPU - epixoip - 06-05-2017
You don't even need benchmarks, it's a matter of very simple maths:
Ryzen7 1800X - 3600 Mhz, 8 cores, 2-way SMT, 128-bit registers == 3600000 * 8 * 2 * 4 == 230400000 32-bit IOPS
Core i7-7700K - 4200 Mhz, 4 cores, 2-way SMT, 256-bit registers == 4200000 * 4 * 2 * 8 == 268800000 32-bit IOPS
Core i7-7740X - 4300 Mhz, 4 cores, 2-way SMT, 256-bit registers == 4300000 * 4 * 2 * 8 == 275200000 32-bit IOPS
Core i7-7800X - 3500 Mhz, 6 cores, 2-way SMT, 256-bit registers == 3500000 * 6 * 2 * 8 == 336000000 32-bit IOPS
Core i7-5960X - 3000 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3000000 * 8 * 2 * 8 == 384000000 32-bit IOPS
Core i7-6900K - 3200 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3200000 * 8 * 2 * 8 == 409600000 32-bit IOPS
Core i7-7820X - 3600 Mhz, 8 cores, 2-way SMT, 256-bit registers == 3600000 * 8 * 2 * 8 == 460800000 32-bit IOPS
So it's literally the slowest of all the high-end desktop CPUs. And of course if you go to the workstation/server side of things, AMD can't even begin to compete with the E5-2630v4, let alone shit like the E5-2699v4. But we were only talking about mainstream CPUs so I won't get into those -- but come on man, the 2699v4 is like 6x faster than the 1800X, so let's not act like AMD is doing anything exciting or innovative.
Also re: the CCX interconnect, I can't believe you would tout that as a positive attribute of the chip architecture. For most hash algorithms we keep everything in registers and in cache, we don't use off-chip memory for the hash calculations. This usually means memory type and speed do not play any real role in the actual hash calculation. HOWEVER, because the Infinity Fabric speed is linked to the off-chip memory speed, you MUST use the fastest DDR4 possible and OC the hell out of it when using more than 4 cores to get the most performance out of the chip. That is ABSURD.
RE: Ryzen 1800X CPU with 1080 Ti GPU - NikosD - 06-05-2017
You really like messing things up.
RyZen 1800X can only be compared with Core i7 7700K from all that huge list you posted, just for impression.
ALL the other Intel chips belong to Intel's HEDT family, which is the worst case scenario for anyone to buy.
I mean it has the worst price/ performance ratio compared to RyZen and in terms of raw performance and perf/ price you should compare all that ridiculous CPUs with the AMD's HEDT platform which is RyZen Threadripper.
Threadripper is going to have, really have, up to 16C/32T with quad channel DDR4-3200 support and 64 PCI-e v3.0 lanes across the whole line!
I mean from 10C/20T to 16C/32T all nine Threadripper models will have quad channel DDR4-3200 support and 64 PCI-e v3.0 lanes.
Just as you were wrong regarding VEGA being slower than Nvidia 1080, you are simply wrong here too, since I don't see real figures of real-world apps like hashcat for the comparison I asked:
RyZen 1800X vs Core i7 7700K
And the Infinity Fabric is a great tool in the hands of AMD in order to produce REAL multi or mega-core CPUs like EPYC which is going to be officially released on June 20th with 32C/64T , 8 channel (!) DDR4 memory support and 128 (!) PCI-e v3.0 lanes.
And you can use Infinity Fabric as a high performance - low latency glue to scale perfectly with a second EPYC 32C/64T in order to get a linear performance of 64C/128T on a dual socket system.
As you can understand, it can eat your Xeons for breakfast, as the demos have already shown.
For 2018, AMD has already announced a EPYC 48C/96T CPU obviously using that weapon, called Infinity Fabric.
BTW, VEGA is using Infinity Fabric too and is going to amaze us in use with RyZen APUs
RE: Ryzen 1800X CPU with 1080 Ti GPU - royce - 06-05-2017
OK, so to recap:
* @NikosD stated that Ryzen 7 1800X "is the best mainstream CPU for everything, even cracking."
* @epixoip specifically addressed hashing performance in his reply, by documenting hashing-relevant IOPS.
* NikosD made follow-up arguments about general performance, not about hashing performance.
NikosD, if you have hashing-specific arguments to make, make them. Otherwise, please stop confusing the issue.
RE: Ryzen 1800X CPU with 1080 Ti GPU - epixoip - 06-05-2017
(06-05-2017, 10:34 AM)NikosD Wrote: RyZen 1800X can only be compared with Core i7 7700K from all that huge list you posted, just for impression.
So... Ryzen 1800X is the best for everything, but only if you compare it to that one specific CPU and ignore all of the other CPUs that are better. Got it! Except that the 7700K is still a good 16% faster, and that's of course assuming you have the fastest overclocked DDR4 possible for the 1800X.
(06-05-2017, 10:34 AM)NikosD Wrote: you should compare all that ridiculous CPUs with the AMD's HEDT platform which is RyZen Threadripper.
You just said that the 1800X is best. Now you're saying Threadripper is better. I don't think you know what the word "best" means.
(06-05-2017, 10:34 AM)NikosD Wrote: Just as you were wrong regarding VEGA being slower than Nvidia 1080, you are simply wrong here too
... Vega is slower than the GTX 1080.
(06-05-2017, 10:34 AM)NikosD Wrote: I don't see real figures of real-world apps like hashcat for the comparison I asked:
If I had benchmarks for the 7700K I'd give them to you, but I don't see that anyone has posted any. However:
1. Benchmarks give only a very narrow glimpse at real-world cracking performance, as you are only measuring single hash brute force for a very short period of time (not long enough for power throttling to kick in.) Benchmarks can be very misleading, like when you claim Vega is faster than a GTX 1080. It may benchmark slightly higher, but it is not faster (thanks, PowerTune!)
2. The formula I used above is a portion of the same formula we've used for years to calculate theoretical max performance of nearly all chips, not just in the absence of benchmarks, but also to measure Hashcat's efficiency. If you understand this formula and what the numbers represent, then you will see why this figure is just as acceptable as benchmark results.
(06-05-2017, 10:34 AM)NikosD Wrote: And you can use Infinity Fabric as a high performance - low latency glue to scale perfectly with a second EPYC 32C/64T in order to get a linear performance of 64C/128T on a dual socket system.
... But you said the 1800X is best.