I think Claymore is best for monero's hashrate
Against a tweaked XMR-stak config file, not really. can u share the tweaked config file? Or please share where to change? You have to make 2 threads for each GPU, roughly half the intensity of only one. That made wonders for Vega in cryptonight, but in Ellesmere and Baffin at least (cards I have) it leverages performance with Claymore's, even a bit faster in my experience. Default settings with autodetection in amd.txt XMR-Stak file are very conservative, there is much more performance to be tweaked from there. That's the advantage of Claymore's, you just run it, and it almost certainly runs at near peak performance. However, as I compile XMR-Stak myself with 0% fee and I can get slightly more performance with a more responsive system (important for me) there is a no brainer. |
|
|
|
|
Considering a hashrate of 22 H/s, which pool is better? aro.cool or aropool.com?
I was mining on the "bigger" pool, decided to try aro.cool for small miners, have 63391 submitted shares in 12 hours and still a pending balance of 0.00...
|
|
|
|
|
The mentioned ETH ASIC's don't make a huge difference in terms of hashrate and power consumption from a multi GPU rig.
Graphic cards can compete.
No so with cryptonight ASIC's, those things had tremendous hashrate at low power.
|
|
|
|
|
Registered on Octaex, no confirmation e-mail, sent email to support as they told at the end of registration, nothing.
Not looking good.
|
|
|
|
Yes, sure, 520 H/s on 65w sounds a terrible deal, but, hey, that's how idiots work.
Anyway, I do mine Cryptonight Lite with Ryzen, you get almost 4x the hashrate with same power consumption.
But do tell me how to get 400 H/s on a Xeon e5645 stock, that's the first time I hear it.
200 is more like it....
Perhaps you're running 12 threads and trashing the cache? If so, try limiting it to 6 threads instead of 12. No, 6 threads only (12 Mb Cache) Here is my cpu.txt "cpu_threads_conf" :
[
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 0 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 2 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 4 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 6 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 8 },
{ "low_power_mode" : false, "no_prefetch" : true, "affine_to_cpu" : 10 },
], But what I get is what I see reported for Xeon e5645 on Cryptonight. It's an old CPU, but it support AES, with 6 cores and 12 Mb L3 cache one should expect higher performance. It was a disappointment. |
|
|
|
I think Claymore is best for monero's hashrate
Against a tweaked XMR-stak config file, not really. |
|
|
|
So what is the general consensus on the currently best CPU mineable coin? I know this is subject to change in the near future. Some are saying Monero. I'm still kicking myself for dumping close to 10k of them @ 50 cents each.
Considering cryptonight coins, maybe Aeon is better on CPU. Cryptonight-lite algo is 3-4x faster on CPU's compared to cryptonight, and only ~ 2x faster on video cards. |
|
|
|
|
When will POW be updated to Aeon V7?
|
|
|
|
|
Yes, sure, 520 H/s on 65w sounds a terrible deal, but, hey, that's how idiots work.
Anyway, I do mine Cryptonight Lite with Ryzen, you get almost 4x the hashrate with same power consumption.
But do tell me how to get 400 H/s on a Xeon e5645 stock, that's the first time I hear it.
200 is more like it....
|
|
|
|
Pre-orders of new 2nd gen Ryzen 7 CPU's are saying "20MB Smart Prefetch Cache".
It should be a beast in Cryptonight / Cryptonight lite.
Total L3 Cache 16MB I saw that too, but why do they advertise and sell on pre-orders 20 Mb cache, and 16 Mb for 1700? Could they have unified 4Mb L2 + 16 Mb L3 cache? 2680 20 threads 25MB and price 200$ Yeah sure, and only 100 H/s more and twice more power consumption than my 8 threads 299€ Ryzen 7 1700... Threads and cache aren't everything. I have a Xeon E5645 - 6 core / 12 threads / 12 Mb L3 cache, and it just sucks at cryptonight, 200 H/s... |
|
|
|
Pre-orders of new 2nd gen Ryzen 7 CPU's are saying "20MB Smart Prefetch Cache".
It should be a beast in Cryptonight / Cryptonight lite.
Total L3 Cache 16MB I saw that too, but why do they advertise and sell on pre-orders 20 Mb cache, and 16 Mb for 1700? Could they have unified 4Mb L2 + 16 Mb L3 cache? |
|
|
|
|
Pre-orders of new 2nd gen Ryzen 7 CPU's are saying "20MB Smart Prefetch Cache".
It should be a beast in Cryptonight / Cryptonight lite.
|
|
|
|
So, no love for Baffin at all?
Please try to use -clkernel 2 for Baffin or try to change the -gt values. We will add auto-tuning in the next version (3.0), which will make the fining of optimal parameters easier. Thanks! Using -clkernel 2 and -gt 50 for Baffin card improved performance from ~11 Mh/s to 14+ Mh/s! Back to PhoenixMiner again! |
|
|
|
|
So, no love for Baffin at all?
|
|
|
|
|
Could performance on Baffin (RX 460/560) be improved?
I have a system with a Baffin card and got around 11 Mh/s while Claymore's gives 14 Mh/s.
|
|
|
|
|
Hi there
I added one more Ellesmere card to my rig (4 now) and Phoenix Miner stopped controlling fan speeds.
-tt and -fanmin stopped working.
I tried to use only one value or assigning a value to each card, but it won't work.
Claymore's works as usual.
Any ideas?
|
|
|
|
Could it support AES also? Some not so old CPU's don't support AVX(2), but AES is around for some time. I don't really know if AES is of some use with Arionum algo, but I have an old 6-core Xeon E5645 that could use that. |
|
|
|
Great, but needs work. 3 threads on a i7-7500U gives 4-5 H/s, while with Java miner using AVX2 gives 7-8 H/s. |
|
|
|
Processor Intel Pentium N4200. There is no avx. Approach for mining?
Just use arionum-miner-java.exe instead. |
|
|
|
Could an Arionum miner be more optimized for Ryzen? Using SHA or so?
The great performer I have is a mobile i7 7500U. With 15W / 3 threads / 2 cores / 4 Mb L3 / 3.1 Ghz sustained speed it hashes as high as 8-9 H/s using Java Miner with AVX2. Truly amazing.
My Ryzen gives the aforementioned 15-16 H/s with 14 threads / 8 cores / 16 Mb L3 / 3.2 Ghz sustained speed. A little shallow compared with the tiny 2 core mobile i7 at around same clock frequencies.
Ryzen truly excels in Cryptonight / lite algos.
But something tells me it arionum mining could be more optimized for Ryzen.
I mined a bit with my old laptop and it gave me 1,7 h/s which is twice better than my old desktop which gave me 0,8-1 k/h. Maybe it just works better with mobile CPUs Each miner utilizes memory, so if you have a memory bottleneck compared to your CPU speed, it will make quite a difference to your h/s I don't think so, Laptop has 16 Gb DDR4 2133 and Laptop has 16 Gb DDR4 3200... |
|
|
|
|
Show Posts
