+- 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: Configuration suggestions (/thread-5289.html)
Configuration suggestions - vladimir125 - 02-29-2016
Hello everyone,
First of all, this is my first time building a custom pc, so please forgive me if I did something stupid or noobish.
I'd wish to build my first cpu dedicated to hash cracking, at the moment my plan is just to play with a single 970 card, in the future I could upgrade to 2x 980 cards.
After reading several setups, I wrote down this configuration, what do you think about it?
CPU: Intel Core i5-6400 2.7GHz Quad-Core Processor
Motherboard: Asus H170-PRO ATX LGA1151 Motherboard
Memory: Kingston HyperX Fury Black 8GB (1 x 8GB) DDR4-2133 Memory
Storage: Western Digital Caviar Blue 1TB 3.5" 7200RPM Internal Hard Drive
Video Card: NVIDIA GeForce GTX 970 4GB Video Card
Case: Corsair 100R ATX Mid Tower Case
Power Supply: Corsair VS 550W ATX Power Supply
Finally a little note about reference video cards: here in Europe is almost impossible to find a reference design 970/980 card, what could you suggest? Do you know any non-reference design that is still good enough?
RE: Configuration suggestions - Xanadrel - 02-29-2016
https://hashcat.net/forum/search.php : search "configuration"
> https://hashcat.net/forum/thread-4689.html
Page 2 : https://hashcat.net/forum/thread-4689-post-26555.html#pid26555
RE: Configuration suggestions - Flomac - 03-03-2016
I'd change most of the system and would buy a
- different mainboard depending on what you want to do with it, but surely not this one; too expensive, Realtek NIC etc.
- different CPU; you either use that CPU for cracking, then get the 6500 for a little more; or you don't then get the 6100 for much less
- memory in 2x4GB configuration (or 2x8GB)
- SSD instead of HDD
- different PSU, since 550W is not sufficient for an upgrade to 2xGPUs
RE: Configuration suggestions - vladimir125 - 03-03-2016
Thank you very much!
Can you please suggest a valid motherboard?
Regarding CPU cracking: is that really an option? AFAIK (and I know very little...) GPU always performs better than CPU, even on slow hashes, especially if you use PRINCE.
But again, I could be totally wrong.
RE: Configuration suggestions - epixoip - 03-03-2016
You are wrong, but not totally wrong. Depends on what you're working with.
RE: Configuration suggestions - vladimir125 - 03-04-2016
(03-03-2016, 11:51 PM)epixoip Wrote: You are wrong, but not totally wrong. Depends on what you're working with.
For curiosity sake, any example? I remember I tried once to use CPU to crack bcrypt hashes and the difference was very small. Again, most likely I'll be wrong, since I tried with an EC2 CPU-oriented instance
RE: Configuration suggestions - epixoip - 03-04-2016
Well, the specific part that was wrong was "GPU always performs better than CPU, even on slow hashes". Slow hashes are especially faster on GPU than CPU (until you start involving memory, that is) because the hash algorithm is slow enough that even a few candidates will keep the GPU loaded, so you don't have to give them much work to gain acceleration.
In reality, it's actually fast hashes & anything involving memory that you have to be concerned about. For example, all of the following are faster on CPU than GPU:
- Attacks without amplifier against fast hash (e.g. straight wordlist or pipe against MD5, MD4/NTLM, SHA1, SHA2, SHA3, etc. )
- Attacks with high hit rate (hundreds+ cracks per second) against medium-to-large sized lists (results in too frequent hash table lookups & collisions + large host-device transfers, in addition to being IO bound from writing to potfile, and god help you if you're also using --remove)
- Very large hash lists (steep drop in bitmap/hash table efficiency + taxing the card's memory controller)
- bcrypt (at one point high-end CPUs and GPUs were about equal for single hash, but CPU still won multi-hash and even a handful of bcrypt hashes can make an ASIC hang.)
- scrypt when 128*r*N >~ 1MiB (I'm actually not entirely sure where the precise tipping point is, but N=16384,r=8,p=1 will definitely make an ASIC hang with even a single hash.)
I know there are more examples, but fuck it I'm drunk, so you take what you get AND YOU LIKE IT
Also keep in mind the architecture changes in v2.10 will allow you to use both CPU & GPU simultaneously, so yeah, CPU kind of matters now as it can provide a decent boost in performance -- especially when you have multiple systems clustered, you can get like another GPU's worth of performance out of your CPUs. For example, http://pastebin.com/MY2BrmEA -- that's 6 GH/s NTLM on CPU alone, which is faster than some of the GPUs you guys are still using!
RE: Configuration suggestions - vladimir125 - 03-04-2016
I know there are more examples, but fuck it I'm drunk, so you take what you get AND YOU LIKE IT
Well, I just had a full and clear explanation, so no complains here
OT: You guys should really start a book/ebook blog series about the hidden mechanics under hash cracking. Most of the times I feel like pressing buttons on big black block
RE: Configuration suggestions - Flomac - 03-04-2016
(03-04-2016, 12:15 PM)epixoip Wrote: Also keep in mind the architecture changes in v2.10 will allow you to use both CPU & GPU simultaneously, so yeah, CPU kind of matters now as it can provide a decent boost in performance -- especially when you have multiple systems clustered, you can get like another GPU's worth of performance out of your CPUs. For example, http://pastebin.com/MY2BrmEA -- that's 6 GH/s NTLM on CPU alone, which is faster than some of the GPUs you guys are still using!
I agree with you except of one point: I don't care how fast a single device will perform, I will always break it down to price/performance-ratio. The CPU you mention is 3000 bucks shy, so yeah, maybe it's faster then some GPUs, but it actually costs as much as three Titan X. So you should really know what you're gonna crack and before you throw a bunch of money in a fast CPU you might have better used it in faster/better GPUs.
It's all relative, that's what I'm saying.