forget about cn-gpu crap. It is much better on nvidia. And will be easy for fpga/asic if have any volume
Yup. it's much better on Nvidia but floating point math is not efficient on FPGAs and ASICs. Are you joking? I've read about fpgas doing scientific floating points math years ago )) It is what they are designed for )) |
|
|
|
|
forget about cn-gpu crap. It is much better on nvidia. And will be easy for fpga/asic if have any volume
|
|
|
|
It seems you could with vega 14nm undervolt a lot without losing hashrate, it looks you cant on vega 7nm as much as you could with vega 14nm.
with such wide bus and fast memory all algos start bottleneck to compute power. So, you'll need to keep gpu freq high => high voltage => high power consumption |
|
|
|
Hi, why this miner(gpu mine) don't work on lower cpu very well? on intel q4400 4gb ram 9-gpu(6-rx 580-8gb,3-rx 480-8gb) the cpu is --no-cpu is delete on config and is 100% usage why? on srbminer is working fine low cpu usage
you shouldn't delete --no-cpu option if you wanna your cpu to take rest )) |
|
|
|
|
bminer 14.1 is still better as pool hashrate reported ))
|
|
|
|
|
2 RuMiner
Oh yeah! Almost useless rigs now )) Guess they are 3gb and 4gb card version )) And with weak celeron cpu which cannot hold grin with all of them ))
|
|
|
|
Guys, what results should one expect on TRTL v2 with polaris and vega? I have 8.2khs with rx578 micron 1200/2140@900mV and 15.5 khs with vega 56 hynix 1380/1050@875mV with another miner. Should I expect boost with jce? BTW, my R5 1600@3.8ghz is doing nice 10khs )) Thanks, JCE |
|
|
|
So, after around 19hs mining TURTLECOIN with the new algo, a bit of feedback: Reminder: my PC/rig is composed of Ryzen 5 2400G (12GB DDR4 2400) on a X470 mobo and 4X AMD RX570 4GB with bios mod. My GPU hashrate is 4,22X higher than with the previous algo, which means GPUs are much better off against FPGAs. I have now around 33KH/s, I had at most 7,8KH/s before. My CPU hashrate is 6X higher than previously (I had 860H/s, now 5150H/s) which means to me that the new algo really favors CPUs for better decentralisation. The Ryzen G has little cache (8MB L3) and I use it for display, meaning that I can't access the whole L3 cache. I expect normal Ryzen series to get an even more impressive boost in performance. The new algo is more power hungry, that doesn't bother me at all considering the HUGGGGE boost in performance. With the CN lite V7 it was 450W/H (CPU + GPU), now 570W/H (CPU + GPU) with CN ultra lite. Among my RX570s, one has a much weaker ASIC than the others, so I had to reduce the intensity compared to the other GPUs, otherwise it hangs out after less than 30mn mining. So the weak one intensity is 960, the others including my beloved little Sapphire ITX is 1008. The network fee is ridiculously low (love it  ), 0.03TRTL for a >10K payment by the pool. Thanks to JCE for the amazing miner, thanks to TURTLECOIN devs for really taking care of their mining community  FPGAs will have even more impressive boost with this low-mem algo. Don't worry ))) Only heavy type algos leave hope for gpus |
|
|
|
|
all pci-e t0 Gen2 and enable 4G encoding.
Some pci-e settings in BIOS are located at strange place - in boot menu ))
|
|
|
|
On 1060 6gig Windows 10 will work?
doesn't work atm |
|
|
|
Hi,
I'm using Bminer v13.0 and as soon as I launch the miner, it keeps throwing this error.
"undefined but requested solver: cuckaroo29"
Is there a quick fix? Thank you
download links on the first page of the topic are old. Use download links from miner's site |
|
|
|
25% more performance than Vega64 with same power consumption, this ist still Vega just in 7nm and some features are working now.
it is 7nm Vega with twice wider memory bus and higher clocks for the same wattage. It should give significant performance increase for ethereum and cryptonight variants. Especially for heavy (much mem required) algos. |
|
|
|
|
krnlx/djm version needs more virtual memory than cryptodredge
|
|
|
|
Hello. I have rx470 8gb micron with modded bios from the previous owner. The bios has 1150Mhz 0.875V gpu and 2000 mhz memory. But after installing drivers no need to patch with "atikmdag-patcher". If I modded stock bios with these values I need to patch with "atikmdag-patcher". How it works? I want to edit bios and no patching after
HEX mod probably. I don't understand the big troubles with Pixel patcher   it doesn't work with new adrenalin 2019 drivers. Or maybe new version of patcher already exists? |
|
|
|
|
why donate level is 5% in 0.15.0.6 beta? Help shows it must be 2% by default.
Also before donation mining miner shows almost all "green" percentage of "effort". But after donation mining all "efforts" are red - more than 150%.
It should be some bug I think
|
|
|
|
480 8gb (samsung mod timing) 1200/2000/0.92v
1135 h/s miner srb 1.6.7 Intensity 58
 Are you sure you are talking about Cryptonight Heavy Hashrate? My result is 745 H/s on RX 480, ETH timing mod, clocks 1150/2150. XMR-AEON-STAK-2.6.0 polaris 8gb makes 1100-1200h/s on heavy algo easily vega does 1800-2000h/s |
|
|
|
Exccelent! On b9 speed on my RX 580 8Gb is 1195 h/s. Devfee is 0,9% - so my is about 1185 h/s. But total shares is 158/2/1 on 4:15 hours:min. So 3 from 158 - is about 1,89% bad shares... Effective hasrate is only 1015 h/s - it's low...
SRB gives 1187 h/s. Devfee is 0.85% - so my is 1177 h/s. Now will test - how much bad shares gives SRB...
xmrig is even better than both jce and srb if we talk about poolside performance xmrig didn't support latest drivers. Many compute errors... So it no in concuration... why the hell you need to install latest drivers? It is mining not gaming. Install proper drivers. |
|
|
|
Exccelent! On b9 speed on my RX 580 8Gb is 1195 h/s. Devfee is 0,9% - so my is about 1185 h/s. But total shares is 158/2/1 on 4:15 hours:min. So 3 from 158 - is about 1,89% bad shares... Effective hasrate is only 1015 h/s - it's low...
SRB gives 1187 h/s. Devfee is 0.85% - so my is 1177 h/s. Now will test - how much bad shares gives SRB...
xmrig is even better than both jce and srb if we talk about poolside performance |
|
|
|
|
they need to adopt some kind of heavy-cn8 algo ...
|
|
|
|
It was b6 version and haven algo. There was no big gpus, only rx470 4gb that shows really the same hashrate with r9 290x 08:45:36 | Hashrate GPU Thread 5: 395.44 h/s
08:45:36 | Hashrate GPU Thread 6: 395.14 h/s - Total GPU 2: 790.58 h/s
08:45:36 | Hashrate GPU Thread 7: 785.26 h/s - Total GPU 3: 785.26 h/s
But shares sent and effective hashrate is low for 290x and normal for rx470 It is not diff problem as I had also weak cpu mining along with gpus in this miner windows. It has much lower hashrate but reported and effective values are very close. So I think it is some kind of hardware errors that I cannot see in miner output or some bug in haven (heavy) code for Tahiti/Hawaii I confirmed the same problem on another test rig with vega56, rx570 8gb and r9 390 8gb. While vega and rx570 show correct numbers - hawaii 390 card reported more hashrate than actual one based on submitted shares. First rig was using old Aug23 blockchain drivers. Test rig is using 18.5.1 driver edit: hashing Ryo with low Diff (5000) on my dual HD7950 (totaling 845h/s) and 840h/s effective hashrate so far with the B-x, with 1% bad shares, as expected. I strongly advise you redo your test with either a fixed low Diff or your big GPU paused. Please provide settings of your test so I can try to reproduce with my cards |
|
|
|
|
Show Posts
