Search content
Sort by
Showing 20 of 39 results by 5tr1k3r
Re: AOCLBF (was Phoenix Rising): Front End GUI to Phoenix/POCLBM
Here goes feedback:
I made an insight on temperatures monitoring and found out something.
I have 5770 and 5850. According to GPU-Z (and some other sources), there are three temp sensors in each card. My sensors values for 5770: 71, 83 and 77. For 5850: 76, 82, 81.
Your monitoring seems to show values of the first sensors, which detect the temperature of a display controller. And these are the lowest values of all three.
MSI Afterburner has one general graph for each graphic card, they are called simply GPU1 temp and GPU2 temp. GPU-Z also has additional GPU temp graph for general GPU temp monitoring. I was surprised to know that both programs use different sensors to show general graphic card temperatures. It's the second sensor for 5770 and the first sensor for 5850. So temps they are showing are 83 and 76 respectively.
Anyway, last paragraph was kinda offtopic. I was just trying to elaborate what sensors are best to use. I propose second ones, they are showing the highest temperatures of the graphic cards. I could finally throw away that afterburner temps from my tray.
About drop-down menu after "Pause mining when GPU ..." I happen to have three options for this: 0, second 0 and 3. Though I have two cards o,o
About clocks tab: I didn't quite get why do I have to click "Clock speed" every time I want to apply changes
Same for "Fan speed". What about "Save settings" concept?
I think, "BFI_INT" would be more informative and appropriate than "HD5xxx+".
Huh, looks like I'm done for now
Thanks for impressive work.
Update: oh, sorry, forgot to tell the version, it's 1.7c.
Quote
If the link between OpenCL GPU list and OS Index is correct (if setting an OpenCL Device automatically selects the corresponding GPU correctly)
Yes, it works fine, GPU is correctly selected. Though it might be difficult for somebody to remember core codenames.Quote
If ADL detects the same number of GPUs as OpenCL (if the OpenCL Device list has same number of GPUs as number of tabs for controlling GPU clocks).
Everything's good here too, same amount of tabs. If you're interested, adapter and OS indexes are the same for each card.I made an insight on temperatures monitoring and found out something.
I have 5770 and 5850. According to GPU-Z (and some other sources), there are three temp sensors in each card. My sensors values for 5770: 71, 83 and 77. For 5850: 76, 82, 81.
Your monitoring seems to show values of the first sensors, which detect the temperature of a display controller. And these are the lowest values of all three.
MSI Afterburner has one general graph for each graphic card, they are called simply GPU1 temp and GPU2 temp. GPU-Z also has additional GPU temp graph for general GPU temp monitoring. I was surprised to know that both programs use different sensors to show general graphic card temperatures. It's the second sensor for 5770 and the first sensor for 5850. So temps they are showing are 83 and 76 respectively.
Anyway, last paragraph was kinda offtopic. I was just trying to elaborate what sensors are best to use. I propose second ones, they are showing the highest temperatures of the graphic cards. I could finally throw away that afterburner temps from my tray.
About drop-down menu after "Pause mining when GPU ..." I happen to have three options for this: 0, second 0 and 3. Though I have two cards o,o
About clocks tab: I didn't quite get why do I have to click "Clock speed" every time I want to apply changes
Same for "Fan speed". What about "Save settings" concept?I think, "BFI_INT" would be more informative and appropriate than "HD5xxx+".
Huh, looks like I'm done for now
Thanks for impressive work.Update: oh, sorry, forgot to tell the version, it's 1.7c.
http://up.idiaths.ru/?id=a1659025fb0fe858
Temperatures seem to be wrong, hm. I'll explain a little: as you can see, I have 2 graphic cards, 5770 and 5850. Their adapter indexes are 3 and 0 respectively. But in order to get them working I have to choose 1 as a device name for 5770 and 0 for 5850. When I put 1 and 0 in "Pause mining ..." I get pretty much the same temperatures (as you can also see), which can't be true. I tried 3 for 5770, it shows some 64, while it's supposed to be hotter than 5850. But for 5850 it works good, because it shows same values as msi afterburner (it's down there, in the tray). Maybe it's just a coincidence.
Temperatures seem to be wrong, hm. I'll explain a little: as you can see, I have 2 graphic cards, 5770 and 5850. Their adapter indexes are 3 and 0 respectively. But in order to get them working I have to choose 1 as a device name for 5770 and 0 for 5850. When I put 1 and 0 in "Pause mining ..." I get pretty much the same temperatures (as you can also see), which can't be true. I tried 3 for 5770, it shows some 64, while it's supposed to be hotter than 5850. But for 5850 it works good, because it shows same values as msi afterburner (it's down there, in the tray). Maybe it's just a coincidence.
I guess would be cool to draw some mhash/s chart somewhere in the main window (additional tab maybe), to monitor performance and stuff like pool uptime. I haven't seen this in any miner I used.
Actually, afterburner's gpu temperature monitor works fine, but it would be more cool like this
Also I suggest to change textboxes type such way, so after clicking in there I don't need to erase all the words like "Device name" or "Pool's name" or whatever. The text remains there, but after clicking on textbox it disappears allowing me to write something. Just a minor thing though.
Actually, afterburner's gpu temperature monitor works fine, but it would be more cool like this
Also I suggest to change textboxes type such way, so after clicking in there I don't need to erase all the words like "Device name" or "Pool's name" or whatever. The text remains there, but after clicking on textbox it disappears allowing me to write something. Just a minor thing though.
http://up.idiaths.ru/?id=ae8663b198fad34a
It can't detect temperatures. The rest works fine, thanks. Btw there are afterburner's temps in tray.
It can't detect temperatures. The rest works fine, thanks. Btw there are afterburner's temps in tray.
Re: BTC Guild - 0% Fees, Long polling, SSL, JSON API, and more [~1500 gH/sec]
Hey eleuthria, you know what would really make the site better? If you have a "total reward" that added up the confirmed and unconfirmed. Put it on top of confirmed or below unconfirmed and the page that shows miners and work would be alot more awesome. 
I've just thought about it. Fully agree.http://up.k10x.net/boqwmfhcqizel/PR1.44.rar
I didn't happen to test the new feature, because pool was never down since the beginning of testing.
Kind of can't get the whole logic of this: what happens if I have options "Restart after an idle time of 30 seconds" and "If 2 restarts occur in 20 seconds then switch to backup"?
Will it ever switch?
Seems to me like second time is pretty much unnecessary.
Anyway, good job, thanks.
By the way, is it possible to kill all phoenixes when quiting? I forget to stop them manually sometimes.
I didn't happen to test the new feature, because pool was never down since the beginning of testing.
Kind of can't get the whole logic of this: what happens if I have options "Restart after an idle time of 30 seconds" and "If 2 restarts occur in 20 seconds then switch to backup"?
Will it ever switch?
Seems to me like second time is pretty much unnecessary.
Anyway, good job, thanks.
By the way, is it possible to kill all phoenixes when quiting? I forget to stop them manually sometimes.
Damn, deepbit's down again. I definitely vote for this solution: switch to backup after two-three (or more) periods of idling. For example, if I have "Restart after an idle time of 30 seconds" option, switching will be executed after 1+ minutes of idle time.
If you don't like the idea, then I definitely vote for button that forces a client to switch from primary to backup or backup to primary, it was mentioned before by smackdaddy.
That's just too sad to lose 10 minutes of mining when you can lose only one
If you don't like the idea, then I definitely vote for button that forces a client to switch from primary to backup or backup to primary, it was mentioned before by smackdaddy.
That's just too sad to lose 10 minutes of mining when you can lose only one
http://up.idiaths.ru/?id=b88aea398748f557
Temperatures are not the same as msi afterburner (or rivatuner) is showing. It shows higher values, 74 & 68 respectively. Obviously it uses another gpu sensor. Guess it'd be the right choice for you to use these sensors too
Device numbers are not the right numbers (that are used in settings of your program), they should be 0 and 1, not 0 and 3
Temperatures are not the same as msi afterburner (or rivatuner) is showing. It shows higher values, 74 & 68 respectively. Obviously it uses another gpu sensor. Guess it'd be the right choice for you to use these sensors too
Device numbers are not the right numbers (that are used in settings of your program), they should be 0 and 1, not 0 and 3
Quote
The blinking part is when its redrawing the text. It might be hardware acceleration-prone, not sure. I can turn down the refresh rates a bit so the blinking isn't as severe.
Maybe it would be a nice idea to redraw the text when it actually needs to be redrawn (right now it redraws even without any needs)? When some stats have actually changed accepted/rejected shares counter has increased or mhash/s performance has changed (with the given accuracy, like, 300 became 299 or so).Quote
If miner idles (hashrate~0 or phoenix hasn't sent out text updates, which it does every time it refreshes hashrate/accepted/rejected, etc.) more than the set time and not responsive to pings.
I'll eventually add an option to use x accepted shares in y minutes, unless someone has better ideas.
I guess that the problem is this: a pool is responsive to pings, but however (for any reason) it cannot offer you any mining work, so miner is idle. And your program decides to keep trying to mine here and not making an attempt to switch pools. Am I right on this? I'll eventually add an option to use x accepted shares in y minutes, unless someone has better ideas.
Quote
I did finally figure out how to initialize ADL (finally!).
Well, that's truly great Sorry for almost-quadruple-posting, but this is really important.
How long does it take to switch to the backup pool? I can't figure it out myself.
Considering some pools are experiencing ddos nowadays (especially deepbit), it becomes crucial to switch to another, backup pool at the right time.
Once I noticed that switching happened after one minute of idling. But now it takes much longer for some reason.
I would really appreciate if you, lvlrdka22, let users change this amount of time in settings.
How long does it take to switch to the backup pool? I can't figure it out myself.
Considering some pools are experiencing ddos nowadays (especially deepbit), it becomes crucial to switch to another, backup pool at the right time.
Once I noticed that switching happened after one minute of idling. But now it takes much longer for some reason.
I would really appreciate if you, lvlrdka22, let users change this amount of time in settings.
I think I've discovered the reason why Phoenix Rising has started to put out such low hashrates compared to before. Checking the PR.log file the default argument is now missing the VECTOR flag.
Manually adding VECTOR into the "other flag" field brought hashrate back up to its original levels. If you could have it automatically add that flag in the next version that'd be great. Or a check box?
I think he actually told about it a couple of times Manually adding VECTOR into the "other flag" field brought hashrate back up to its original levels. If you could have it automatically add that flag in the next version that'd be great. Or a check box?
Checkbox's good. Autovectors - not good, I think, because I would want to set all flags by myself.
Oh, and some more: is it possible not to switch focus to the new phoenix window? I mean, can it be done without losing focus to the application I'm currently in? Sorry for weird words, don't know how to correctly say that.
Wow, what a great tool! Thank you!
Two things:
There are some moments when OSD is constantly blinking almost every second (no matter is click-through on or off), which is kinda annoying. Didn't figure out what causes this. Windows 7, aero disabled.
I'm sure you know that device finder doesn't really show the correct device numbers
And minor things:
It would be good if password was stored like this: "*************".
Maybe would be also good to let the user to hide OSD when he needs to. Like, some checkbox "show OSD" or something.
Desktop lag is more significant compared to the usual phoenix with the same settings.
Two things:
There are some moments when OSD is constantly blinking almost every second (no matter is click-through on or off), which is kinda annoying. Didn't figure out what causes this. Windows 7, aero disabled.
I'm sure you know that device finder doesn't really show the correct device numbers
And minor things:
It would be good if password was stored like this: "*************".
Maybe would be also good to let the user to hide OSD when he needs to. Like, some checkbox "show OSD" or something.
Desktop lag is more significant compared to the usual phoenix with the same settings.