mersenneforum.org Navi (RX 5700, RX 5700XT)
 Register FAQ Search Today's Posts Mark Forums Read

2019-07-23, 20:48   #23
lavalamp

Oct 2007
London, UK

22×3×109 Posts

Quote:
 Originally Posted by SELROC so what ?
I would expect the prices to drop slightly some time after the launch, surprising to me that you would buy it before you can use it.

2019-07-24, 03:01   #24
SELROC

61·103 Posts

Quote:
 Originally Posted by lavalamp I would expect the prices to drop slightly some time after the launch, surprising to me that you would buy it before you can use it.

what happened with the RX580 when miners started buying them in quantities?

the price jumped high, and only after long time the price dropped.

2019-09-26, 09:15   #25
snme2pm1

"Graham uses ISO 8601"
Mar 2014
AU, Sydney

241 Posts

Quote:
 Originally Posted by xx005fs I am not expecting much from these cards
Are there any further indications for the recent AMD 7nm devices?

Last fiddled with by snme2pm1 on 2019-09-26 at 09:41 Reason: broken format

 2019-10-08, 09:25 #26 Lorenzo     Aug 2010 Republic of Belarus 2·5·17 Posts Hello! Are there some benchmarks for 5700's cards?
 2019-10-29, 17:54 #27 James Heinrich     "James Heinrich" May 2004 ex-Northern Ontario 55 Posts If anyone has benchmarks for any of the Radeon RX 5700-series cards I would very much appreciate them, please submit here: https://www.mersenne.ca/mfaktc.php#benchmark
2019-10-31, 18:44   #28
Mark Rose

"/X\(‘-‘)/X\"
Jan 2013

23·359 Posts

Quote:
 Originally Posted by James Heinrich If anyone has benchmarks for any of the Radeon RX 5700-series cards I would very much appreciate them, please submit here: https://www.mersenne.ca/mfaktc.php#benchmark
Would adding test to the Phoronix Test Suite be useful? Michael at Phoronix generally ends up with most of the cards, so perhaps he'd be willing to run a trial factoring test, especially if it were easy and automated.

 2019-10-31, 22:24 #29 James Heinrich     "James Heinrich" May 2004 ex-Northern Ontario 55 Posts I don't know anything about Phoronix, but if they could run a mfakto/mfatkc benchmark as part of their suite that would probably provide a lot of useful data points. I don't know how much realtime they'd want to dedicate to such a test, but perhaps suggest an exponent with no known factors in the chosen range, perhaps something like Code: Factor=4294967189,76,77 I picked the largest exponent < 232 that has no small factors (it's been tested to 86), since larger exponents will allow testing higher bit kernels in a reasonable amount of time. The above should take roughly 10 mins on an old slow card like my RX480, which means current new cards should complete in a few minutes and get less as technology improves. Don't want the runtime to become ridiculously short in the near future, as that makes for a less accurate measurement. I'll leave it with you if you want to suggest it as a possible benchmark, and you can tell me when the data starts flowing in.
 2019-11-28, 14:00 #30 nomead     "Sam Laur" Dec 2018 Turku, Finland 23·41 Posts SETI@Home having problems: Flakey AMD/ATI GPUs, including RX 5700 XT, Cross Validating, polluting the Database The thread started way back in August already, but the problem seems to be getting worse over time, with more cards participating. OpenCL problems on Navi / RDNA?