So is it a FPGA or ASIC? Or it has gender identity crisis?
It should be FPGA since it support 2 algos. Just like other Baikals. According to girl, its ASIC. Also, it doesnt support 2 algos, cryptonight-lite is just cryptonight, but with half of the scratchpad. |
|
|
|
I recently decided to take another look at RAM timings now that many people have been playing with custom timings for several months.
Instead of zeroing FAW/32AW as I do in my strapmod utility, I think it is better to set them low. 7/10 seems to be standard for Hynix straps, while 6/8 is typical for Samsung. My testing so far using 6/8 on Hynix gives the same performance for eth mining as 0/0, while potentially being more stable at higher memory clocks.
I also suspect the field labels may be reversed, so FAW is really 32AW, and vice-verse. In my testing, when I set "FAW" to 8 or 10 while leaving "32AW" to zero, I see no slowdown in ethash speed.
I may update my strapmod when I'm done, but my goal is to finally finish a Linux tool I started working on last year that tweaks timings at runtime. This allows for tuning the memory timing for ethash, equihash, or cryptonight without reflashing the BIOS and rebooting. It has already greatly improved strap testing, as I have been able to modify timings on the fly while mining; I don't even have to restart the miner. I still crash the GPU a lot, as dynamically switching between two sets of good timing will still cause a hang depending on which timings you switch in what order.
I have sort of been under the assumption that in the presence of an invalid value, memory controllers must be calculating t32aw automatically. Every strap I've ever seen (including from mfgs) is out of spec, given that t32aw should always be >= 8 x tfaw, as i understand it. Btw, would love to take a peek at your tool  If my suspicion about the labels being reversed is correct, and 32AW is a multiplier for FAW, then 6/8 is perfectly fine. FAW would be 6*RRD, and 32AW would be 8*6*RRD. That also would better explain the observation that setting what they think is 32AW to zero causes instability. As for the tool I'm working on, you can get an idea of how it works by looking at my fork of amdmeminfo. https://github.com/nerdralph/amdmeminfoI modified it to show runtime strap values. My tool doesn't just read them, it writes them too. I believe Yurio is ahead of me in this process with his Windows miner now being able to do runtime timing modifications. https://github.com/zawawawa/GatelessGateSharpThat formula seems wrong, or i am missing something. Elpida is as follow: 1500: 6/10/7 1625: 7/12/8 1750: 7/12/8 2000: 7/12/8 No matter how you put it, neither tFAW and tFAW32 can be any multiple of tRRD. Currently using tFAW = 0 and tFAW32 = 4. For what i remember actually, tFAW if its set to 0, its actually read as 8. Dunno if each single value counts as 8, or if its just timing properties. If its the first - that would make sense for the 1625-2000Mhz straps. That makes possible the following : tFAW = 7 * 8, Which would also fall in line with what you said. |
|
|
|
I had found a new better memory strap for Elpida RAM. More than 1kh/s cryptonight xmr !  RX 470 : 1020 h/s , sgminer @1250 core, @2100 mem, 66W (gpu-z) 942 h/s , @1150 core, @2050 mem, 52W (gpu-z) 777000000000000022AA1C00AC615B3CA0550F142C8C1506006004007C041420CA8980A9020004C 01712262B612B3715 I donīt know the eth hash speed.  Maybe somebody can check it ? This strap is bad. Both of my cards bsod hard with it. Also its slower and runs with more power. Trfx and ras2ras are too tight. I often see TRFC and ras2ras for elpida straps at that value Its actually default value for 1500MHz strap Make it better My bad, i didnt say that properly. The trfc and ras2ras for xryptonight is nearly useless, you lose about 10h/s when going to 150 for example. But the power cons drops by 5watts. Thats for cryptonight. Also, all trcd values are not properly done, the A value should be the same as the non A one. Make it better ! I test so much straps, and that are the best, I can find. Okay... ####SEQ_RAS_TIMING#### TRCDW = 12 TRCDWA = 13 TRCDR = 24 TRCDRA = 22 TRRD = 5 TRC = 60 ####SEQ_CAS_TIMING#### TNOPW = 0 TNOPR = 0 TR2W = 26 TCCDL = 2 TCCDS = 5 TW2R = 15 TCL = 20 ####SEQ_MISC_TIMING#### TRP_WRA = 44 TRP_RDA = 24 TRP = 22 TRFC = 97 ####SEQ_MISC_TIMING2#### PA2RDATA = 0 PA2WDATA = 0 TFAW = 0 TCRCRL = 3 TCRCWL = 4 T32AW = 0 TWDATATR = 0 ####ARB_DRAM_TIMING#### ACTRD = 23 ACTWR = 18 RASMACTRD = 38 RASMACTWR = 43 ####ARB_DRAM_TIMING2#### RAS2RAS = 97 RP = 43 WRPLUSRP = 55 BUS_TURN = 21 This is yours. Below is yours, with the fixes applied. ####SEQ_RAS_TIMING#### TRCDW = 12 TRCDWA = 12 TRCDR = 24 TRCDRA = 24 TRRD = 5 TRC = 60 ####SEQ_CAS_TIMING#### TNOPW = 0 TNOPR = 0 TR2W = 26 TCCDL = 2 TCCDS = 5 TW2R = 15 TCL = 20 ####SEQ_MISC_TIMING#### TRP_WRA = 44 TRP_RDA = 24 TRP = 22 TRFC = 97 ####SEQ_MISC_TIMING2#### PA2RDATA = 0 PA2WDATA = 0 TFAW = 0 TCRCRL = 2 TCRCWL = 2 T32AW = 4 TWDATATR = 0 ####ARB_DRAM_TIMING#### ACTRD = 23 ACTWR = 18 RASMACTRD = 38 RASMACTWR = 43 ####ARB_DRAM_TIMING2#### RAS2RAS = 97 RP = 43 WRPLUSRP = 55 BUS_TURN = 21 See the difference? TRFC and RAS2RAS are left like that, since you use the same strap for Ethash. Otherwise, you increase them by 50% for Cryptonight. I already gave you critic, now i even give you an example of how it would better in terms of stability/speed. |
|
|
|
Heh, got my Elpida strap stable.
At 1200/2000Mhz, i am getting 975h/s with 896/8/2 and Compute Mode.
At 1280/2000Mhz - 1020h/s.
With the other card, i am getting 1068h/s at 1380/2000, or 1038h/s at 1280/2000.
Can you share your straps? its not finished, not 100% stable on both cards. I dont have enough free time to update it.
Another tweak I've wondered about is monitoring memory die temps. GDDR5 chips have internal thermal diodes similar to the GPU, and they can be enabled by setting a bit in mode register 7. Knowing the temperature of each memory die could provide useful diagnostic info for problems such as a heatsink not having good contact with the chip.
Yes! Mem temp management is critical to getting Vegas stable - I'm sure it would be hugely useful for stabilizing polaris as well. I now use a rule of thumb of core generally being about 10š cooler than mem, but clearly this is a very rough estimate, and depends significantly on heat sink design (or lack thereof.) Also might give some sense of the 'asic quality' of the mem (if that's a thing...) From personal experience, TFAW = 0 is not an issue, but TFAW32 = 0 alongside, causes instability, often seen with Monitor Sleep and background refreshes. Using TFAW32 = 4. I recently decided to take another look at RAM timings now that many people have been playing with custom timings for several months.
Instead of zeroing FAW/32AW as I do in my strapmod utility, I think it is better to set them low. 7/10 seems to be standard for Hynix straps, while 6/8 is typical for Samsung. My testing so far using 6/8 on Hynix gives the same performance for eth mining as 0/0, while potentially being more stable at higher memory clocks.
I also suspect the field labels may be reversed, so FAW is really 32AW, and vice-verse. In my testing, when I set "FAW" to 8 or 10 while leaving "32AW" to zero, I see no slowdown in ethash speed.
I may update my strapmod when I'm done, but my goal is to finally finish a Linux tool I started working on last year that tweaks timings at runtime. This allows for tuning the memory timing for ethash, equihash, or cryptonight without reflashing the BIOS and rebooting. It has already greatly improved strap testing, as I have been able to modify timings on the fly while mining; I don't even have to restart the miner. I still crash the GPU a lot, as dynamically switching between two sets of good timing will still cause a hang depending on which timings you switch in what order.
I have sort of been under the assumption that in the presence of an invalid value, memory controllers must be calculating t32aw automatically. Every strap I've ever seen (including from mfgs) is out of spec, given that t32aw should always be >= 8 x tfaw, as i understand it. Btw, would love to take a peek at your tool Well, 8*0 is 0 There is probably something like that as well, but there is clearly a difference for me, higher instability when both are 0. |
|
|
|
I had found a new better memory strap for Elpida RAM. More than 1kh/s cryptonight xmr ! RX 470 : 1020 h/s , sgminer @1250 core, @2100 mem, 66W (gpu-z) 942 h/s , @1150 core, @2050 mem, 52W (gpu-z) 777000000000000022AA1C00AC615B3CA0550F142C8C1506006004007C041420CA8980A9020004C 01712262B612B3715 I donīt know the eth hash speed. Maybe somebody can check it ? This strap is bad. Both of my cards bsod hard with it. Also its slower and runs with more power. Trfx and ras2ras are too tight. I often see TRFC and ras2ras for elpida straps at that value Its actually default value for 1500MHz strap My bad, i didnt say that properly. The trfc and ras2ras for xryptonight is nearly useless, you lose about 10h/s when going to 150 for example. But the power cons drops by 5watts. Thats for cryptonight. Also, all trcd values are not properly done, the A value should be the same as the non A one. |
|
|
|
I had found a new better memory strap for Elpida RAM. More than 1kh/s cryptonight xmr ! RX 470 : 1020 h/s , sgminer @1250 core, @2100 mem, 66W (gpu-z) 942 h/s , @1150 core, @2050 mem, 52W (gpu-z) 777000000000000022AA1C00AC615B3CA0550F142C8C1506006004007C041420CA8980A9020004C 01712262B612B3715 I donīt know the eth hash speed. Maybe somebody can check it ? This strap is bad. Both of my cards bsod hard with it. Also its slower and runs with more power. Trfx and ras2ras are too tight. |
|
|
|
|
Heh, got my Elpida strap stable.
At 1200/2000Mhz, i am getting 975h/s with 896/8/2 and Compute Mode.
At 1280/2000Mhz - 1020h/s.
With the other card, i am getting 1068h/s at 1380/2000, or 1038h/s at 1280/2000.
|
|
|
|
|
For R9 290, its seems more limited by core. Increasing core from 945 to 1050mhz gave me 850h/s with stock memory straps at 1500mhz mem. Also, this was way back, when the compute drivers still werent released.
|
|
|
|
i need a good cryptonight strap for h5gq8h24mjr
I need it too. Don't we all? Akakakakakkakaa, count me in. For 1150 2100, max i can get only 930 h/s for xmr. How about others? which card is used to get 930 amd or nvidia My RX 470 4 GB Elpida Card makes 930 H/s XMR ... Core @1150, Mem @ 2050 ... sgminer 896/8/2 ... 55W(gpuz) ... 0 Memoryerrors R0landīs Strap mix: 777000000000000022AA1C00315A5436A0550F15B68C1506004082007C041420CA8980A9020004C 01712262B612B3715 (copy from PBE) The data from GPUZ are a little bit different : 777000000022AA1C0315A5436A055F15B68C1560408207C41420CA8980A9204C01712262B612B37 15 Why are the data different ? It skips some double 00 pairs and shows them as either 1 zero or doesn't show them at all. Also, it moves the bits around. This is not raw format. Not in endian, the bits are probably grouped together for each timing. Also, i dont see how you can be making 930h/s with that strap at such low clocks and this wrong miner config. At 1380/2000, with the same strap and 896/8/2, sgminer barely pulls 500h/s. With 896/8/1 its doing 700h/s. With 512/4/2 its doing 850h/s. Also, that strap TRCDR is not the same with TRCDRA. Same miner config for my RX470 and RX480 cards gives me even lower hash rates (450h/s). Considering, you need to do ~1k raw intensity total on both Linux and Windows, i dont see how you can be pushing 1.8k raw intensity. Pretty much depends...seeing this strap I seriously doubt it can handle 2050 without errors. If the card is dedicated, the strap works realy good and stable (24h, 0 rejected shares). But if you also use the card for working with the OS, sometimes the miner doesnīt start with 896 worksize or BSOD. Dedicated no problems. At 1100/2000 it makes ~30 ETH wth claymore single mode, and no error. Donīt try 1380/2000 ... 1100/2000 @850mv is maybe better (~890h/s) . Good luck  Okay.... I found the issue. It was the drivers. The way, you, Roland were doing things were entirely different. By default, the RX *70 and *80 cards maximum Raw Intensity is 512 at -w 4 and 2 threads. BUT, the Compute mode (BlockChain driver), is the thing that helps with overcoming this issue. I tested today with 17.12.11 with Gaming mode, Compute mode, the mentioned strap, and sgminer configs of 512/4/2 and 896/8/2. Results were as follow: 235w at wall (whole rig, testing only the 580, so assume ~100w for card, ~80% PSU eff.). For all straps and configs. Temps ... around the same. 1380mhz Core/2000mhz Mem, 1100mv/750mv. Roland strap, 512/4/2, gaming mode : 750h/s. Roland strap, 512/4/2, compute mode : 750h/s. Roland strap, 896/8/2, gaming mode : ~700h/s. Roland strap, 896/8/2, compute mode : 950h/s. My strap, 512/4/2, gaming mode : 950h/s. My strap, 512/4/2, compute mode : 950h/s. My strap, 896/8/2, gaming mode : ~750h/s. My strap, 896/8/2, compute mode : 1024h/s. did you test it for eth other than that 30MH test? i don't understand why so many zero's went away -_- What ETH? I think you got the wrong quote. Anyway, if you provide me with proper sgminer config, i will test it. I tested with : -X 448 -w 256 -g 2 at 1380mhz/2000mhz, and it was doing 28.7Mh/s at ~130w for the GPU. If the config is not correct - those numbers will go up. |
|
|
|
i need a good cryptonight strap for h5gq8h24mjr
I need it too. Don't we all? Akakakakakkakaa, count me in. For 1150 2100, max i can get only 930 h/s for xmr. How about others? which card is used to get 930 amd or nvidia My RX 470 4 GB Elpida Card makes 930 H/s XMR ... Core @1150, Mem @ 2050 ... sgminer 896/8/2 ... 55W(gpuz) ... 0 Memoryerrors R0landīs Strap mix: 777000000000000022AA1C00315A5436A0550F15B68C1506004082007C041420CA8980A9020004C 01712262B612B3715 (copy from PBE) The data from GPUZ are a little bit different : 777000000022AA1C0315A5436A055F15B68C1560408207C41420CA8980A9204C01712262B612B37 15 Why are the data different ? It skips some double 00 pairs and shows them as either 1 zero or doesn't show them at all. Also, it moves the bits around. This is not raw format. Not in endian, the bits are probably grouped together for each timing. Also, i dont see how you can be making 930h/s with that strap at such low clocks and this wrong miner config. At 1380/2000, with the same strap and 896/8/2, sgminer barely pulls 500h/s. With 896/8/1 its doing 700h/s. With 512/4/2 its doing 850h/s. Also, that strap TRCDR is not the same with TRCDRA. Same miner config for my RX470 and RX480 cards gives me even lower hash rates (450h/s). Considering, you need to do ~1k raw intensity total on both Linux and Windows, i dont see how you can be pushing 1.8k raw intensity. Pretty much depends...seeing this strap I seriously doubt it can handle 2050 without errors. If the card is dedicated, the strap works realy good and stable (24h, 0 rejected shares). But if you also use the card for working with the OS, sometimes the miner doesnīt start with 896 worksize or BSOD. Dedicated no problems. At 1100/2000 it makes ~30 ETH wth claymore single mode, and no error. Donīt try 1380/2000 ... 1100/2000 @850mv is maybe better (~890h/s) . Good luck Okay.... I found the issue. It was the drivers. The way, you, Roland were doing things were entirely different. By default, the RX *70 and *80 cards maximum Raw Intensity is 512 at -w 4 and 2 threads. BUT, the Compute mode (BlockChain driver), is the thing that helps with overcoming this issue. I tested today with 17.12.11 with Gaming mode, Compute mode, the mentioned strap, and sgminer configs of 512/4/2 and 896/8/2. Results were as follow: 235w at wall (whole rig, testing only the 580, so assume ~100w for card, ~80% PSU eff.). For all straps and configs. Temps ... around the same. 1380mhz Core/2000mhz Mem, 1100mv/750mv. Roland strap, 512/4/2, gaming mode : 750h/s. Roland strap, 512/4/2, compute mode : 750h/s. Roland strap, 896/8/2, gaming mode : ~700h/s. Roland strap, 896/8/2, compute mode : 950h/s. My strap, 512/4/2, gaming mode : 950h/s. My strap, 512/4/2, compute mode : 950h/s. My strap, 896/8/2, gaming mode : ~750h/s. My strap, 896/8/2, compute mode : 1024h/s. did you test it for eth other than that 30MH test? i don't understand why so many zero's went away -_- What ETH? I think you got the wrong quote. |
|
|
|
i need a good cryptonight strap for h5gq8h24mjr
I need it too. Don't we all? Akakakakakkakaa, count me in. For 1150 2100, max i can get only 930 h/s for xmr. How about others? which card is used to get 930 amd or nvidia My RX 470 4 GB Elpida Card makes 930 H/s XMR ... Core @1150, Mem @ 2050 ... sgminer 896/8/2 ... 55W(gpuz) ... 0 Memoryerrors R0landīs Strap mix: 777000000000000022AA1C00315A5436A0550F15B68C1506004082007C041420CA8980A9020004C 01712262B612B3715 (copy from PBE) The data from GPUZ are a little bit different : 777000000022AA1C0315A5436A055F15B68C1560408207C41420CA8980A9204C01712262B612B37 15 Why are the data different ? It skips some double 00 pairs and shows them as either 1 zero or doesn't show them at all. Also, it moves the bits around. This is not raw format. Not in endian, the bits are probably grouped together for each timing. Also, i dont see how you can be making 930h/s with that strap at such low clocks and this wrong miner config. At 1380/2000, with the same strap and 896/8/2, sgminer barely pulls 500h/s. With 896/8/1 its doing 700h/s. With 512/4/2 its doing 850h/s. Also, that strap TRCDR is not the same with TRCDRA. Same miner config for my RX470 and RX480 cards gives me even lower hash rates (450h/s). Considering, you need to do ~1k raw intensity total on both Linux and Windows, i dont see how you can be pushing 1.8k raw intensity. Pretty much depends...seeing this strap I seriously doubt it can handle 2050 without errors. If the card is dedicated, the strap works realy good and stable (24h, 0 rejected shares). But if you also use the card for working with the OS, sometimes the miner doesnīt start with 896 worksize or BSOD. Dedicated no problems. At 1100/2000 it makes ~30 ETH wth claymore single mode, and no error. Donīt try 1380/2000 ... 1100/2000 @850mv is maybe better (~890h/s) . Good luck Okay.... I found the issue. It was the drivers. The way, you, Roland were doing things were entirely different. By default, the RX *70 and *80 cards maximum Raw Intensity is 512 at -w 4 and 2 threads. BUT, the Compute mode (BlockChain driver), is the thing that helps with overcoming this issue. I tested today with 17.12.11 with Gaming mode, Compute mode, the mentioned strap, and sgminer configs of 512/4/2 and 896/8/2. Results were as follow: 235w at wall (whole rig, testing only the 580, so assume ~100w for card, ~80% PSU eff.). For all straps and configs. Temps ... around the same. 1380mhz Core/2000mhz Mem, 1100mv/750mv. Roland strap, 512/4/2, gaming mode : 750h/s. Roland strap, 512/4/2, compute mode : 750h/s. Roland strap, 896/8/2, gaming mode : ~700h/s. Roland strap, 896/8/2, compute mode : 950h/s. My strap, 512/4/2, gaming mode : 950h/s. My strap, 512/4/2, compute mode : 950h/s. My strap, 896/8/2, gaming mode : ~750h/s. My strap, 896/8/2, compute mode : 1024h/s. |
|
|
|
i need a good cryptonight strap for h5gq8h24mjr
I need it too. Don't we all? Akakakakakkakaa, count me in. For 1150 2100, max i can get only 930 h/s for xmr. How about others? which card is used to get 930 amd or nvidia My RX 470 4 GB Elpida Card makes 930 H/s XMR ... Core @1150, Mem @ 2050 ... sgminer 896/8/2 ... 55W(gpuz) ... 0 Memoryerrors R0landīs Strap mix: 777000000000000022AA1C00315A5436A0550F15B68C1506004082007C041420CA8980A9020004C 01712262B612B3715 (copy from PBE) The data from GPUZ are a little bit different : 777000000022AA1C0315A5436A055F15B68C1560408207C41420CA8980A9204C01712262B612B37 15 Why are the data different ? It skips some double 00 pairs and shows them as either 1 zero or doesn't show them at all. Also, it moves the bits around. This is not raw format. Not in endian, the bits are probably grouped together for each timing. Also, i dont see how you can be making 930h/s with that strap at such low clocks and this wrong miner config. At 1380/2000, with the same strap and 896/8/2, sgminer barely pulls 500h/s. With 896/8/1 its doing 700h/s. With 512/4/2 its doing 850h/s. Also, that strap TRCDR is not the same with TRCDRA. Same miner config for my RX470 and RX480 cards gives me even lower hash rates (450h/s). Considering, you need to do ~1k raw intensity total on both Linux and Windows, i dont see how you can be pushing 1.8k raw intensity. |
|
|
|
|
Okay, Elpida is, hands down for me, the worst memory there is. Doing a STABLE and FAST memory strap for XMR is harder than getting a girlfriend for a virgin nerd sucker.
Even the free strap, which was found here (with little changes to RP/TRFC and TRCDR), it still crashes, while giving measly 840h/s.
|
|
|
|
Okay, one thing i find interesting, is the lack of multiple threads. Unlike sgminer, you can't request the miner to run 2 threads on a single GPU. On sgminer and on Windows, this actually brought extra h/s.
Ofc this uses multiple threads There is no way to get 2kh/s out of vega by single thread. Interface simply shows only the total hashrate. Or you probl knew this alrd, but then again, its optimized only for vega, since OP previously only had vega cards. Yea i knew that, still decided to test it. And yea, its testd on RX480/580. I said this, because, it really shows up only 1 thread working - GPU0 and nothing else. No idea if its the same for Vega. |
|
|
|
|
Okay, one thing i find interesting, is the lack of multiple threads. Unlike sgminer, you can't request the miner to run 2 threads on a single GPU. On sgminer and on Windows, this actually brought extra h/s.
|
|
|
|
|
Guys, have you run into issues with Elpida memory, where strap X is working fine from 2000 to ~2100Mhz on 1 card, but on another card it crashes RIGHT away on 2000Mhz. Thats on sgminer and XMR, and both cards are MSI Gaming X RX 580 4GB.
|
|
|
|
|
Since i am getting 900H/s with the RX480 Hynix, it is possible. Its just matter of fiddling, which i didn't really had the time or nerves to do again.
|
|
|
|
|
Okay, i havent seen any Elpida straps for XMR here actually.
Got a 580 4GB Gaming X with Elpida. Did some mixed Samsung/Hynix/Elpida, getting at 1355c/2000m 948h/s.
But, for some reason, even after relaxing the timings, going to 2100Mhz BSODs me instantly with Thread_STUCK.
Thats with TRC=64 and TCL=22.
Is there difference in the timings for the 5xx series, specifically for the
TCRCRL, TFAW/32, ACTRD/WR, RASMACT, RP, WRPLUSRP and BUS_TURN?
Those are the only values, that i didnt bother raising any further from the Samsung/Hynix ones.
Oh, and also, TCKE = 10, which is less than the RX 470 and 480.
Everything else looks completely normal to me.
|
|
|
|
|
The blockchain driver is useless for me for XMR, tried it like 3 different times with DDU, each time i was getting less hash for more power draw.
Also, @doktor83, i will fiddle with this today again, i will install a 580 as well.
BTW, my cards are using Hynix (the 470 and 480), which are running the public strap from this thread, but modified a bit more.
|
|
|
|
896 w8 g2 is the best you can get for 470/480/570/580
and yes, he has some nice straps.
and yes, you can use blockchain drivers for xmr, i get 925mhs on 1130/2000 with sapphire nitro+ 570 4gb
Either you are running it differently, either something is wrong. 896 w8 g2 makes the cards barely hashing, like 50h/s or 400h/s. 896 w8 makes them to hash with 100h/s less than normal. Anyway, for some reason, the blockchain driver on my 470 and 480 give ~20h/s LESS and it increases the power consumption. Go figure. What GPU that you use? And can you try sgminer or gatelessgate? After I check it for few days at pool, the hashrate really increase. sgminer-gm 5-5-5, Windows 10, 17.2.1 drivers, RX470 XFX single fan 1225c/2000m ; RX480 Sapphire Nitro+, 1305c/2000m. Using on both ri 512, w4 g2 and getting 470 = 867h/s, 480 = 900h/s. |
|
|
|
|
Show Posts
