+1 i7 3820 at 4ghz has more than enough oomph, but still 9 was the best.
Maybe with a dual xeon beast 11 will work. (my dream setup, dual 12 core xeon, asus 7 pci-e slot equipped server board and a ton of ram lol...oh and 7 r9 280x's for this miner)
Topic: Primecoin GPU miner: 9.1 CPD on a 280x / pool / 2% dev.fee - page 39. (Read 100962 times)
im using 14.3 but it mines after that as shown also runing cpu usage 11 and look at the usage
Ok, please keep me updated.Re CPU usage:
Please make sure that it's "-cpuusage=11", not "-cpuusage 11" (You'll probably need a lower value if it's the case)
right yes i know
im using 14.3 but it mines after that as shown also runing cpu usage 11 and look at the usage
Try "-cpuusage=9" seems to best setting for most people, including myself.
im using 14.3 but it mines after that as shown also runing cpu usage 11 and look at the usage
Ok, please keep me updated.Re CPU usage:
Please make sure that it's "-cpuusage=11", not "-cpuusage 11" (You'll probably need a lower value if it's the case)
im using 14.3 but it mines after that as shown also runing cpu usage 11 and look at the usage
and this is the error i get when i first start the server with 2 290x's
Never has anything like this before. Also you are first to report it.You mentioned that this is a first start - what heppens after that? Does it look like it's mining? Does it show this error on the next start?
What version of Ctalyst do you use?
1 7970 and a phenom 555
2 7970's and a 8320
2 290x's with 2 6core opterons
and this is the error i get when i first start the server with 2 290x's
2 7970's and a 8320
2 290x's with 2 6core opterons
and this is the error i get when i first start the server with 2 290x's
I am jealous of those 280x numbers. Hoping for a 290x improvement soon....
*************************************************************
*************************************************************
on 4930k with 3 x 290x I increased cpu usage to 11 and it maxed out cpu at 100%. I found the sweet spot at 9. 10 was too much also.
I upgraded to 1.03 and changed that setting and performance went up 10%. Also all 3 cards have similar times now, where previously they were different for each card.
I upgraded to 1.03 and changed that setting and performance went up 10%. Also all 3 cards have similar times now, where previously they were different for each card.
cpuusage=11 results in me getting way slower perf on my i7 3820 at 4ghz...and dual 280x cards.
HT on or off makes no difference.
cpuusage=9 is the sweetspot for me. Insane perf boost, time is 170 per card or so, and total cpd 3.1 with two r9 280x cards.
HT on or off makes no difference.
cpuusage=9 is the sweetspot for me. Insane perf boost, time is 170 per card or so, and total cpd 3.1 with two r9 280x cards.
"-cpuusage=9" is the best setting for me too. It uses 57%~63%.
Est Avg CPD: 1.41
Est Spot CPD: 1.47
time: 192~200
is it normal to get error while first running the server on 290x also while waiting for the wallet to sync my 2 opterons hit a block lol
I just run with Pentium G3220 wih 5x R9 280X.
I got 4.3 cpd a bit slower. Please recommend cpu for use with 5x card per rig.
I got 4.3 cpd a bit slower. Please recommend cpu for use with 5x card per rig.
on 4930k with 3 x 290x I increased cpu usage to 11 and it maxed out cpu at 100%. I found the sweet spot at 9. 10 was too much also.
I upgraded to 1.03 and changed that setting and performance went up 10%. Also all 3 cards have similar times now, where previously they were different for each card.
I upgraded to 1.03 and changed that setting and performance went up 10%. Also all 3 cards have similar times now, where previously they were different for each card.
cpuusage=11 results in me getting way slower perf on my i7 3820 at 4ghz...and dual 280x cards.
HT on or off makes no difference.
cpuusage=9 is the sweetspot for me. Insane perf boost, time is 170 per card or so, and total cpd 3.1 with two r9 280x cards.
HT on or off makes no difference.
cpuusage=9 is the sweetspot for me. Insane perf boost, time is 170 per card or so, and total cpd 3.1 with two r9 280x cards.
I'm using 1.04 and it's showing cpu usage as low as 7% with the "-cpuusage 11", but spikes to 34% some times. Getting time: 240~251 and Est. CPD: 1.17.
...
Could the "-cpuusage" command be buggy?
Please make sure it's "-cpuusage=11", not "-cpuusage 11"....
Could the "-cpuusage" command be buggy?
Just a little side note, the 1.04 still says 1.03.
It's because it's 1.03. We haven't released 1.04 yet.Oh ok, forgot the "=". I'll give it a try....
I'm using 1.04 and it's showing cpu usage as low as 7% with the "-cpuusage 11", but spikes to 34% some times. Getting time: 240~251 and Est. CPD: 1.17.
...
Could the "-cpuusage" command be buggy?
Please make sure it's "-cpuusage=11", not "-cpuusage 11"....
Could the "-cpuusage" command be buggy?
Just a little side note, the 1.04 still says 1.03.
It's because it's 1.03. We haven't released 1.04 yet. New version of the server 1.03 is available on the website.
The same link: http://primegpu.com/files/prime-gpu-server.zip
What's new:
- "-cpuusage" is extended up to 11 with the default value at 6.
Higher values should increase speed on 1- / 2-cards setups and/or with high-end CPUs up to 15%.
Note that you should try to keep CPU working time at 50-70%. So do not set 11 on slow CPUs or multi-card rigs
- 'Jumping mining time' issue is fixed
- Spot CPD output is added
- Up to 8 cards (or 4 7990) are now supported per rig (earlier it was 6 max)
The same link: http://primegpu.com/files/prime-gpu-server.zip
What's new:
- "-cpuusage" is extended up to 11 with the default value at 6.
Higher values should increase speed on 1- / 2-cards setups and/or with high-end CPUs up to 15%.
Note that you should try to keep CPU working time at 50-70%. So do not set 11 on slow CPUs or multi-card rigs
- 'Jumping mining time' issue is fixed
- Spot CPD output is added
- Up to 8 cards (or 4 7990) are now supported per rig (earlier it was 6 max)
I'm using 1.04 and it's showing cpu usage as low as 7% with the "-cpuusage 11", but spikes to 34% some times. Getting time: 240~251 and Est. CPD: 1.17.
With 1.03, I was at 20% cpu usage. Getting time: 225~230 and Est. CPD: 1.29.
I'm running an AMD Phenom II x4 810 (PassMark Score= 3481) with just one MSI Twin Frozr III HD 7970 OC BE.
Could the "-cpuusage" command be buggy?
Just a little side note, the 1.04 still says 1.03.
I opted to get a 2nd license +1 to devs for quick turn around on license request happily running latest build on 2 280x now 2.89 cpd =) the new cpuusage increase has helped
btw i notice the cpd goes to shit if the cpu usage spills over onto ht cores (will try disabling ht from bios when i get a chance)
btw i notice the cpd goes to shit if the cpu usage spills over onto ht cores (will try disabling ht from bios when i get a chance)
What cpu?
I'll try with and without ht on my 3820 then...or use affinity. :p
I opted to get a 2nd license +1 to devs for quick turn around on license request happily running latest build on 2 280x now 2.89 cpd =) the new cpuusage increase has helped
btw i notice the cpd goes to shit if the cpu usage spills over onto ht cores (will try disabling ht from bios when i get a chance)
btw i notice the cpd goes to shit if the cpu usage spills over onto ht cores (will try disabling ht from bios when i get a chance)
Quote
Just want to point out to everyone that I did receive my refund as promised. I haven't found a block yet, but I more than respect the decision to reduce the price of the licenses (given the competition) and issue refunds to those that originally paid 65 XPM.
Some people have not got their payouts yet. Of course we will process everybody's request.
ATM we are trying to process new orders first as well as new licenses just to let people start mining.
There is practically a queue of requests that we do not have time to manage on time.
Feel free to remind us is something is forgotten.
New version of the server 1.03 is available on the website.
The same link: http://primegpu.com/files/prime-gpu-server.zip
What's new:
- "-cpuusage" is extended up to 11 with the default value at 6.
Higher values should increase speed on 1- / 2-cards setups and/or with high-end CPUs up to 15%.
Note that you should try to keep CPU working time at 50-70%. So do not set 11 on slow CPUs or multi-card rigs
- 'Jumping mining time' issue is fixed
- Spot CPD output is added
- Up to 8 cards (or 4 7990) are now supported per rig (earlier it was 6 max)
The same link: http://primegpu.com/files/prime-gpu-server.zip
What's new:
- "-cpuusage" is extended up to 11 with the default value at 6.
Higher values should increase speed on 1- / 2-cards setups and/or with high-end CPUs up to 15%.
Note that you should try to keep CPU working time at 50-70%. So do not set 11 on slow CPUs or multi-card rigs
- 'Jumping mining time' issue is fixed
- Spot CPD output is added
- Up to 8 cards (or 4 7990) are now supported per rig (earlier it was 6 max)
Jump to: