Show Posts
|
Pages: [1] 2 3 4 »
|
there is no 9 yet, you downloaded a malware delete it
what a qualified contribution 
|
|
|
I think this coin is already dying slowly 
|
|
|
I have an issue with the amount of RAM allocated. I used TRM miner with the following command: eth_big_mode_adjust=64 after I changed my mainboard, this value 64 wont work anymore. it sais that it cant allocate that much ram, it allocates around 7900 MB of memory. I have a AMD 5700 and was able to allocate all but 64 MB of video ram, but after I switched my mainboard, there needs to be around 256 MB of free memory that cant be allocated by TRM. Must be the mainboard / bios...I searched all values but didnt find anything that I can change for Vram reservation. any idea? costs a lot of mhash 
|
|
|
question for RTM setup:
Wrong place, no RTM support here. where s the right place?
|
|
|
question for RTM setup: whats the impact of ram for the rtm hashrate? more mhz is as important as low latency or more important or is latency more important? 3600 better than 2666, sure cl13 better than cl16, sure, but what about: 3600 cl16 vs 2666 cl13 e.g so whats more important, mhz or latency? big impact on price  cpu is ryzen 9 5900
|
|
|
so more mhz is as important as low latency?
3600 cl16 better than 2666 cl13 e.g.?
|
|
|
Im setting up a cpu mining rig too atm. I read that fast ram timing help to get more hashrate, so cl14 is better than cl16. but what about the frequency. is 2666 mhz faster than 2133 e.g.? or is the most important the timing?
|
|
|
I use TRM with the following command: --eth_big_mode_adjust=64 worked fine for month. I needed to change my MB and now recognized, that there is an error message at the beginning of the mining process. sth. like: not able to allocate all dag size of 8000 MB, only able to allocate 7900 MB, may result in lower hashrate. for some reason Iīm not able to allocate all except 64MB of video ram anymore. what or whoīs blocking my video ram? could that be the new MB? Iīm on AMD now with a MSI B450 GAMING PLUS MAX (MS-7B86) with AMD Ryzen 3 1200. Might be a setting in BIOS? pls help.....loosing some mhash 
|
|
|
i saw many 5500xt that can do around 29mh at 50 watts. Maybe someone have one for me for tests?  It looks like that this is data from HiveOS or RaveOS, it isn`t correct. You need to use a wattmeter to get true values with AMD GPUs. my better 5500xt do 27 mh at 53-55 watts.
my lessor ones do 25 mh at 51 watts
I only do memory straps
I never flash bios.
My 5500xt are about 28mh at 75-80watts, only RaveOS configured. My 5600xt are flashed and about 43mh at 100-110watts. Before flashing 5600xt gave me about 39-40mh with the same power consumption. Yeah here goes. all smos Msi 5600xt 37.93 mh at 79 watts Msi 5600xt 37.92 mh at 78 watts Msi 5600xt 37.93 mh at 78 watts Msi 5600xt 38.77 mh at 77 watts Msi 5600xt 37.95 mh at 76 watts Msi 5600xt 38.17 mh at 91 watts Msi 5600xt 38.17 mh at 86 watts Msi 5600xt 38.39 mh at 79 watts powercolor 5700xt 49.66 mh at 106 watts powercolor 5700xt 49.60 mh at 103 watts powercolor 5700xt 49.58 mh at 106 watts powerColor 5700xt 50.88 mh at 113 watts------ an rma replacement card Xfx 5700xt 50.36 mh at 100 watts amd 5700xl 49.48 mh at 103 watts amd 5700 xl 49.54 mh at 104 watts msi 5700xl 49.45 mh at 109 watts ------- this card has issues msi 5700xt 50.28 mh at 115 watts ------- this card also has issues msi 5700xt 40.49 mh at 95 watts -------- this card really is weird asus 5500xt 26.02 mh at 61 watts msi. 5500xt 26.02 mh at 56 watts msi 5500xt 25.44 mh at 56 watts asrock 5500xt 24.67 mh at 72 watts ------ a mini card runs hot also a replacement rma his 5500xt 25.19 mh at 56 watts ------ good mini card his. 5500 xt 25.29 mh at 57 watts------ good mini card of the 24 above the 5600xt clearly do best those are the smos software numbers for watts and mh This Power consumption data is totally useless. YOU NEED TO MEASURE POWER AT WALL Card does not show the total power needed...only Nvidia cards can show good power data, AMD always shows less. My 5700 do 55 Mhash at around 60 watts....so thatīs the real efficiency king   At wall this rig makes 550 W....so its 2 times more in total.
|
|
|
Nvidia just announced that they will add SAM (smart access memory) to new drivers soon. This should work with all CPUīs.....not like the feature works with AMD, where only the 5000 series CPU is supported. So if you can use SAM with all CPUīs, will you take advantage of this new feature and allow 2 mining instances? this could double the hashrate at the 10gb cards like the 3080. or do you concentrate on AMD support? and did you already use SAM for AMD with 5000 series CPUīs?
|
|
|
after a day of testing....I now see some savings in power as I was able to reduce core voltage down to 730 and it seems to be stable. so if you look from this side...good job TRM team !!! I think my fans rotate at lower speed too....so less memory heat...but this cant be verified....just a feeling 
|
|
|
I installed the new version and tested the new kernel.
I can see huge changes to 7.22 e.g. I could reduce core clock by 125 mhz but the reduction did not save any power off the wall. seems like any higher core clock is cut in mhash and power. when I reach the sweetspot and get the same mhash than before (with now 140 mhz less core clock, slightly before max mhash) I need the same power off the wall. I also reduced core voltage from 755 to 750 what gives me some watts out of the wall, but the power consumption at the same mhash is the same than in 7.22. so for me 7.22 was the big difference, not 8.0.0 btw, Iīm Using a bios modded asus 5700 in hiveos with 20.30 drivers.
I think I now can go lower than 750 with core voltage what could save some more watts for sure.....that would be great. I can see that you reach same mhash with less core clock....but what counts for me is watt / mhash and thatīs nearly the same (at the same voltage settings).
what about the memory allocation? do you use the huge free vram of the 8gb cards as long as eth dag size is that low and fits almost twice into the ram? (phoenix has a kernel called turbo where he can load dag file twice into ram)
|
|
|
temp of the msi card is regulated via gpu temp...as mem temps go much higher while mining, the standard temp management is not useable while mining  unfortunately you cant read mem temps with nvidia cards, you only see if temp is too high if hashrate drops and card is throtteling thatīs why you choose a custom fan speed in miner-parameters or use a custom fan curve. but i want to find a setting that I can use for both, mining and gaming
|
|
|
Hey guys. I use my gaming machine for mining. MSI card start fans at 60 °C, thatīs sometimes a little bit too high for me...I can play some games in 0-fan mode...So I set custom fan curve. Minimum fan speed is 30 % which is absolutely noiseless. Thatīs the better option I think...playing games with 30 % instead of 0-fan. But in custom fan curve mode you cant set a fixed fanspeed in miner. And as the miner does not raise gpu temp that much, my custom curve would result in mining with 30 / 35 % which is too low for mining. So I could shoose autofan which is less fan for gaming (in my opinion) or custom wher I needed to set the fan too agressive to raise it above 40 while mining  any suggestions for that situation? strange
|
|
|
hmm...
my information is that internal power management of the new nvidia cards reduces not only gpu power, it also reduces ram, so below 50 % memory power drops significantly, independent from core clock / power. So my information is that you need at least around 50 % power target to mine, independent from max tdp....thatīs why I asked....did anyone here already try a 3060 / 3070 with higher tdp?
|
|
|
Hey guys. I have the option to buy some of these cards: https://www.msi.com/Graphics-Card/GeForce-RTX-3070-SUPRIM-8GThe Suprim variant has a TDP of 270W, the normal one has 240 W which can be found here: https://www.msi.com/Graphics-Card/GeForce-RTX-3070-GAMING-TRIOFor almost all 3060 / 3070 normally you put 50 % TDP for mining to have a good hash / wattage ratio. Most of these cards have 220 / 240W TDP. Who knows what hashrates can be expected when I put the Suprim to 50 % TDP (135W) or 44 % TDP (120W). I suggest that it would mine the same hashrate when I set it at 50 %, so it would consume 135 instead of 120 W for mining the same hashrate than the other MSI 3070 card. I assume that putting on 44 % TDP would consume same power than other cards (120W) but would result in less hashrate. Could that be correct? Is the Suprim variant not the best choice for mining? Thx guys.
|
|
|
that would work. but I start / stop it several times a day. I would love to see a solution that calculates the uptime automatically, without manual adding the times. I could write start and stop times into a txt file and then automate the calculation of week / monthly uptime...but Iīm not a programmer, I cant do a small program by myself, thatīs why I asked if thereīs some solution thatīs already available.
awesomeminer logs the total uptime and you can show the total per month. but ONLY for the payed version of awesomeminer. the pool logs uptime for each miner, but you can only show charts when your miners are online, you have no addition of the total uptime.
argl...
|
|
|
As I need Windows for my gaming machine, I cant use a mining os  Phoenix shows itīs uptime too when itīs running. Bbut when I close the miner, play a game and then restart the miner, it resets itīs count to 0. So I would need to write down the time every time. Thatīs not what Iīm looking for.
|
|
|
I do not want to measure the power ! I want to measure the time the phoenixminer.exe runs. Thatīs 2 different things.
I dont know why you always suggest hardware for measuring power from the wall.
Itīs time....measured in hours or days that I need. For only 1 program, not the whole uptime of the pc. The time the miner runs per month.....300 hours in december, 500 hours in november. thatīs what I want to know. Hope you now understand my question.
Thx
|
|
|
the powermeter would measure electricity for gaming and mining. So no solution for what I asked for 
|
|
|
|