|
Very nice finding!
BTW I managed to reach 51.6mh, but in CN-R I can't get past 2150 2180h/s no matter what i try.
What is the trick there?
|
|
|
|
However I am not seeing the same for the other formula RAS = RCD + WR
For example for Vega 64 defaults... RAS(33) = RCDRD(16) + RCDWR(10)... nope.
Sure it's good. RAS = MAX(RCDRD,RCDWR) + WR ! Anyone got a tool for displaying memory errors under linux? (other than trm) |
|
|
|
Have you managed to find a good value for RAS to go with the low RCD?
OK... I'm revealing my secrets.  Here are the formulas straight from wikipedia, not hard to find.  If you check ras, rcd, wr, rp, rc you can clearly see the formulas were used. These values are not hocus pocus. They are calculated according to specific rules. So if you change ras, you should change rc too and vica-versa |
|
|
|
So, someone done with 51 Mh ETH  True ? How about Windows ? It's already in this thread... with exact values. |
|
|
|
I'm also stable with the same timings and --rfc 250 which gave me an additional boost. Modified timings would be --rp 10 --rc 44 --rfc 250 --rrd 3 --rcdrd 12 --rcdwr 5 Thank you for the info. I will give it another go. BTW for current version '--rrd 3' should be replaced with '--rrds 3 --rrdl 3' otherwise it won't be set Edit: Tried it again on my ref V64 Samsung, but same... it crashes after about 1 min. BTW what's the point in such low rcd if you don't set ras accordingly? |
|
|
|
Eliovp, great work on this tool. Here are some of my results:
OS: linux/Ubuntu 18.04
miner: teamredminer 0.4.3
card: Vega 64 air cooled (MSI air boost OC w/ samsung mem)
core clock: ~1150Mhz
mem clock: 1100Mhz
timings modified: --rp 10 --rc 44 --rfc 300 --rrd 3 --rcdrd 12 --rcdwr 5
hashrate w/ stock timings: ~19.5kh/s
hashrate w/ new timings: 21.36kh/s
Unfortunately I don't have power readings from this rig, so I'm not sure how much the power usage increased with the new timings.
I think the card can be pushed more, but I don't have the time to keep tuning at the moment.
Holy OMG !!!!!! I dont understand the various parameters, but i just followed todxx timings on my vega 64(air cooled/samsung). Results: Miner : SRBMiner 1.8.3 Algo : Cryptonight litev7 Clocks : 1408/1107/875 (ODT) Actual : 1354/1107/850 (HWinfo) Stock Timing : 4530 h/s Todxx Timing : 4800 h/s Thx Eliovp for the tool and todxx for making it way easier. Will keep posting the magic as i try more algorithms Is that working stable for anybody? that rcdwr is very low. GPU crashes in a few minutes for me. |
|
|
|
I download "WinAMDTweak.exe" and "EIO.dll" form Releases, after I running WinAMDTweak.exe --current, it show "Failed to load EIO.dll", any wrong with my step?
You need those 4 files for win I guess... |
|
|
|
(PATH to folder where MemTweak is stored)\WinAMDTweak.exe --CL 16 --RAS 30 --RCDRD 14 --RCDWR 14 --RC 44 --RP 14s --RRDS 3 --RRDL 5 --RTP 5 --FAW 20 --CWL 7 --WTRS 4 --WTRL 9 --WR 16 --WRRD 1 --RDWR 19 --REF 17550 --MRD 8 --MOD 15 --PD 8 --CKSRE 10 --CKSRX 10 --RFC 244
Adjusting CL does not affect hashrate directly for me at all. As higher CL is more stable, I'd leave it at stock 20. Another remark is that you set RCDWR the same as RCDRD... it gets faster if you decrease it. But the max I could squeeze out from it is 51.5 @ 1100MHz with my mods, without it it was a bit less. Edit: I'm on 18.6.1, maybe 19.3.3 is faster? Edit2: FAW=20 does not work for me. it's unstable. |
|
|
|
on cnr managed to achieve 2.5 kh/s ? on polaris? Doubt it... @yobigd20 Don't you forget what's said in the readme. Set the timings during load for cards that has different timings in different memory powerstates. I assume you set the timings in idle state. |
|
|
|
Tell me pls, who managed to achieve some better results for rx570-580 using this tweak? Are there concrete examples of what needs to be changed and where, I personally did not understand, read all the help, but in order to specifically change something I did not dare  . It's already been said that you won't see any benefits with this tool for cards that have public optimized timings like all the RX 4xx/5xx (micron/elpida/hynix/samsung). Of course if you want to further optimize timings this tool comes handy, but you need background knowledge. |
|
|
|
|
Testing stability the past two days.
The light timing with ETH 51mh is rock solid.
The more edgy timings for CN algos needs to be smoothened because after a few hours mining stopped. It was 2130h/s till then.
BTW I tried kernel 4.16 but totally same behaviour lik 4.15. The main problem is I can't set p7 above 1700MHz to get ~1500 at low voltage.
|
|
|
|
|
There is a tool for mem timings strap encode/decode. Google it.
But for polaris the public straps are quiet good. They are included in PBE. You can also see them in the source.
Anyone willing to share results of Radeon VII ? I'm curious.
If 110mh is possible with stock timings, then there is much potential in the memory. Might be core bottlenecked.
|
|
|
|
|
Keep in mind that some cards got very hot if external power is on, but mobo/riser is powered off. It can consume 10-20W without fan moving and can fry itself after some time.
If you power riser and pci-e power from the same psu, then you avoid this situation.
|
|
|
|
Today I tried ethash, but disappointment. I can only reach 48.0 MH/s with Phoenix For ethash it should be tuned towards high bandwidth and not low latency Update: Started tuning all over from a different approach and got 51+ MH/s This timing seems stable so far. But got lower CN hashrate than the previous (~2050). Update2: 51mh is verified by the pool side 6hr average and no invalid shares.  Hint: Lowest stable RAS for me is 30. You can get the other numbers with the formula. Any hint from the experts what to do with all the mysterius params at the lower part of the table?  |
|
|
|
my results with Vega56 (samsung) on CN-R so far  What miner are you using to get those figures? I'll be honest I'm a little disappointed. I get around 2050 per Vega56@64 with no memory timing tweaks using Teamredminer at 1458 core clock (~1410 effective), pretty much like pbfarmer. ...or is the miner slower on Linux? That is XMRigCC-AMD Teamredminer might be faster... Which is the fastest CN miner for Linux? I use SRB for win. |
|
|
|
|
1600MHz core needed to drive out tuned 1100MHz HBM2 on the 2 CN forks I tested so far.
Daggerhashimoto is different in this sense.
|
|
|
|
you’re prob running 225-250w for that 2100h/s, or about 80-100w more than you would need for 2000, without timing mods (I got 2K @ 837mv.) +50% power for +5% h/r is not worth it - even if your power is free!
Normally 1500MHz needs around 925mV. I don't get those sub 850 numbers. Do you have a watt meter on it? Vddci is also set to <837? to do the math in my case: 975/925=1.054 and 1.054^2 = 1.11 (+11%), and considering the higher clock, that means power consumption is only ~15% higher then before. @coinscrow:Thank you mate, going to try that! |
|
|
|
could you share your values? and also is your vega 56 flashed to the vega 64 bios or stock?
Basically started with your settings and then added some more. I'd like to enjoy the race a bit longer so I don't share it directly but I give you a hint: you can find some simple formulas quiet easily that help calculating the numbers.  Vega 64 bios, yes. |
|
|
|
Holy shit, that's nice man.
Is it stable? Is this ubuntu or you use hiveos/minerstat/similar?
Ubuntu 18.04.2 LTS and AMD-gpu-pro 18.50 Still struggling with 4.15 kernel not able to set clocks properly. Anything above 1700MHz in P7 state puts the GPU in crimp mode. So I have to use 1700/1030mV in P7 to get the highest possible coreclk. Ideal would be something like 1800/975mV in P7, that would make ~1650 effective clock in P7 at much lower wattage I learned I will have to upgrade kernel, but don't know which version, there are also complains everywhere about 4.17-4.20 all over the forums. I'm currently testing stability. Wondering what would be the eth hashrate with these timings. Implying on CN algos it gives +15%, 45mh*1.15=51.75mh Gonna test it later. |
|
|
|
my results with Vega56 (samsung) on CN-R so far |
|
|
|
|
Show Posts
