rx 560 vs 1050 vs 1050ti
#11
(12-07-2017, 11:28 PM)Vidramon Wrote: @tacohashcat I will love to see real world of 1050 !
Since I think that this guy on YT video is untrue.

My r9 270x benchmarks 96064 H/s and in real world it do 93500 H/s whics is almost the same. So I love to see that 1050 but I think he is not in right.

Also I will try, to get some benchmark results from rx560, I don't know does I will succeed but I will try.
I'll run a WPA hash for you this evening when I get home and post up the results. Again, that machine isn't a cracking rig, so the card normally just drives the display. If I were to be using hashcat regularly, I would try to go with a 1070FE or higher. But if budget is a concern, then whatever you can afford.
Reply
#12
Also @Vidramon, I use this site if I am looking for certain cards. I am not sure if there is a version where you are at, but check it out http://www.nowinstock.net/computers/videocards/nvidia/
Reply
#13
Using Hashcat 4.0.1 (no beta, whatever is on the main page is what I used) on Windows 10, workload 3, GeForce drivers 388.59, and MSI afterburner push the GPU fan up to 100% and running against the sample hashcat WPA/WPA2 hccapx:

Session..........: hashcat
Status...........: Running
Hash.Type........: WPA/WPA2
Hash.Target......: 8381533406003807685881523 (AP:ae:f5:0f:22:80:1c STA:98:7b:dc:f9:f9:50)
Time.Started.....: Fri Dec 08 17:32:04 2017 (9 mins, 46 secs)
Time.Estimated...: Thu Jan 09 02:15:28 2020 (2 years, 31 days)
Guess.Mask.......: ?1?2?2?2?2?2?2?3 [8]
Guess.Charset....: -1 ?l?d?u, -2 ?l?d, -3 ?l?d*!$@_, -4 Undefined
Guess.Queue......: 1/8 (12.50%)
Speed.Dev.#1.....:    84117 H/s (56.45ms)
Recovered........: 0/1 (0.00%) Digests, 0/1 (0.00%) Salts
Progress.........: 45342720/5533380698112 (0.00%)
Rejected.........: 0/45342720 (0.00%)
Restore.Point....: 696320/89248075776 (0.00%)
Candidates.#1....: E2s8cerd -> Eq6wfane
HWMon.Dev.#1.....: Temp: 54c Fan:100% Util: 99% Core:1695MHz Mem:3504MHz Bus:16

[s]tatus [p]ause [r]esume [b]ypass [c]heckpoint [q]uit =>

I never saw it go above 84kH/s, but I didn't update the status too often. The card reached 99% at that capture, so I wouldn't expect it to top 85kH/s in real world applications. As you can see this was after almost 10 mins running, so it had plenty of time to get going. You could try workload w4, but I didn't want to do this since the card I am using pushes display to my monitor and it may become unresponsive.

I never saw temps go above 55C on the card. This is a small GPU with one fan and an OEM cooler, not FE. Power comes from the motherboard, no PCI-E power connectors needed. No other cards are around to block air to the GPU, and my case has good airflow.

Edit: This is also running whatever the stock clocks are. Afterburner would allow tweaking the overclocking, so you might be able to push a bit extra out of the card.
Edit2: and of course, this is on my GTX 1050.
Reply
#14
-w4 is mandatory for highest performance since Hashcat 4.
Reply
#15
(12-08-2017, 04:21 AM)Flomac Wrote: I did indeed. Does not change the fact. You seems to be on a very tight budget, otherwise my recommendation would be a GTX1060. The 2GB GTX 1050 is available for ~100€ in Germany, for that price the pp-ratio is ok. At 165€ for a GTX1050 Ti it isn't.

1050 ti already go over budget but if it is better than 1050 and it is I can borrow some money from friend to buy it since no one buy graphic cards often. This buy will stay couple of years with me for sure, but 1050 ti is limit as my money go for graphic card.

(12-09-2017, 01:31 AM)tacohashcat Wrote: Using Hashcat 4.0.1 (no beta, whatever is on the main page is what I used) on Windows 10, workload 3, GeForce drivers 388.59, and MSI afterburner push the GPU fan up to 100% and running against the sample hashcat WPA/WPA2 hccapx:

Session..........: hashcat
Status...........: Running
Hash.Type........: WPA/WPA2
Hash.Target......: 8381533406003807685881523 (AP:ae:f5:0f:22:80:1c STA:98:7b:dc:f9:f9:50)
Time.Started.....: Fri Dec 08 17:32:04 2017 (9 mins, 46 secs)
Time.Estimated...: Thu Jan 09 02:15:28 2020 (2 years, 31 days)
Guess.Mask.......: ?1?2?2?2?2?2?2?3 [8]
Guess.Charset....: -1 ?l?d?u, -2 ?l?d, -3 ?l?d*!$@_, -4 Undefined
Guess.Queue......: 1/8 (12.50%)
Speed.Dev.#1.....:    84117 H/s (56.45ms)
Recovered........: 0/1 (0.00%) Digests, 0/1 (0.00%) Salts
Progress.........: 45342720/5533380698112 (0.00%)
Rejected.........: 0/45342720 (0.00%)
Restore.Point....: 696320/89248075776 (0.00%)
Candidates.#1....: E2s8cerd -> Eq6wfane
HWMon.Dev.#1.....: Temp: 54c Fan:100% Util: 99% Core:1695MHz Mem:3504MHz Bus:16

[s]tatus [p]ause [r]esume [b]ypass [c]heckpoint [q]uit =>

I never saw it go above 84kH/s, but I didn't update the status too often. The card reached 99% at that capture, so I wouldn't expect it to top 85kH/s in real world applications. As you can see this was after almost 10 mins running, so it had plenty of time to get going. You could try workload w4, but I didn't want to do this since the card I am using pushes display to my monitor and it may become unresponsive.

I never saw temps go above 55C on the card. This is a small GPU with one fan and an OEM cooler, not FE. Power comes from the motherboard, no PCI-E power connectors needed. No other cards are around to block air to the GPU, and my case has good airflow.

Edit: This is also running whatever the stock clocks are. Afterburner would allow tweaking the overclocking, so you might be able to push a bit extra out of the card.
Edit2: and of course, this is on my GTX 1050.

Hmm does not fit !
Your benchmarks shows 94 kH/s and real is 84 kH/s.
Mine sold r9 270x benchmarks shows 96 kH/s and in real it do 93 kH/s, using -w 3 and windows 7 Hashcat 3.5.0.

1050 is more efficient since it consume 75w relative to r9 270x 180w but I was hoping your benchmarks will be close to real time cracking. I found https://hashcat.net/forum/thread-6423.html this is 115 kH/s and this one https://hashcat.net/forum/thread-6167.html 120 kH/s.

So I can expect from your experience 10 kH/s lower which means it is about 105 kH/s 1050 ti.
Don't know but as far as I see it is only choice for me since no benchmarks for rx560 and I wont to take a risk.
Reply
#16
(12-08-2017, 03:39 AM)epixoip Wrote:
(12-07-2017, 10:42 PM)tacohashcat Wrote: I'm not sure that epixoip works with lower end cards

Sorry, I don't. GTX 1070 is the lowest I'll go and even then I stopped offering it in the Brutalis as soon as the 1080Ti came out, making the GTX 1080 our lowest-end offering.

I was hoping for another "f*** AMD" rant Big Grin How long till we see Titan Vs in a Brutalis box?
Reply
#17
The Titan V as far as I have seen, is only available direct from Nvidia right now, and is limited to two. Might take a bit to be able to get a decent quantity, especially since a Brutalis alone takes 8.
Reply
#18
@Vidramon,
I ran the same command in -w4 mode. Was worried I might have a serious lag on the IO, but no noticeable lag was seen.

Session..........: hashcat
Status...........: Running
Hash.Type........: WPA/WPA2
Hash.Target......: 8381533406003807685881523 (AP:ae:f5:0f:22:80:1c STA:98:7b:dc:f9:f9:50)
Time.Started.....: Sat Dec 09 07:35:48 2017 (3 mins, 9 secs)
Time.Estimated...: Thu Sep 26 07:47:53 2019 (1 year, 290 days)
Guess.Mask.......: ?1?2?2?2?2?2?2?3 [8]
Guess.Charset....: -1 ?l?d?u, -2 ?l?d, -3 ?l?d*!$@_, -4 Undefined
Guess.Queue......: 1/8 (12.50%)
Speed.Dev.#1.....:    97633 H/s (372.22ms)
Recovered........: 0/1 (0.00%) Digests, 0/1 (0.00%) Salts
Progress.........: 18022400/5533380698112 (0.00%)
Rejected.........: 0/18022400 (0.00%)
Restore.Point....: 0/89248075776 (0.00%)
Candidates.#1....: 3arierin -> 3ssrpine
HWMon.Dev.#1.....: Temp: 59c Fan:100% Util:100% Core:1695MHz Mem:3504MHz Bus:16

As you can see I was up over 97kH/s. I was averaging 97.5-98kH/s, although at one point at the beginning I saw 103kH/s Smile
Temps I never saw go above 60C at 100% fan with stock clocks as per before. So yes, the 1050 will get the same performance as your 270x. If you tweak it a bit, I am sure you could get over 100kH/s easily, while still staying in the safety of power consumption.
Edit: a quick change of the clock speeds, and I was in the 99kH/s range. I only added 25MHz to get that.
Reply
#19
(12-09-2017, 02:10 AM)Flomac Wrote: -w4 is mandatory for highest performance since Hashcat 4.
You are correct, thank you. I went back and ran again with -w4, and was averaging 97-98kH/s, which is spot on with the benchmark.
Reply
#20
Those threads you posted @Vidramon were for older versions of hashcat, so take those numbers with a grain of salt. I think with a little bit of tweaking, you could get a 1050 to get damn close to the numbers of a 1050Ti. So in my eyes, I would skip the Ti. Not worth the extra money. I would get a 1050, or save up and keep looking for deals on a 1060. Not sure when nvidia plans on announcing more volta cards, but when they do, that will push pascal prices down. Be ready to pounce when they go clearance.
Reply