OFFICIAL CGMINER mining software thread for linux/win/osx/mips/arm/r-pi 4.11.1

[os2sam]

soryy about the post --temp-cutoff 75
it was just for testing purpose to see when cgminer cuts off gpu and if it will starts again when temperature drops.

--temp-target <arg> Target temperature when automatically managing fan and GPU speeds (default: 75)
--temp-overheat <arg> Overheat temperature when automatically managing fan and GPU speeds (default: 85)
--temp-cutoff <arg> Temperature where a device will be automatically disabled, one value or comma separated list (default: 95)

As I said I think the temp cutoff requires manual interaction whereas the temp overheat is where it recovers from.

So are you working yet?

yes it works ok. thank you
but as I said maybe developer will think about more auto setup cgminer which will lower gpu engine with fan settings or temperature settings

cut off doesnt restart gpu

Yes, that would be a great new feature.  I wonder why he never thought of that.

Anyway I'm glad your running without burning up your GPU(s).
Good Luck,
Sam

[1714324687]

1714324687

[1714324687]

1714324687

[Blisk]

I think that is not so hard to do for developer. I don't know nothing about it so just assume.
For example for gpu fan speed
you take that cgminer measuer top mhas from 245mh to 252mhashes
and fan speed from 3450 to 3500.
so cgminer know that if you have 350mhashes must be a gpu engine arround 850 and it takes a little lower as it is 800
and than if I set gpu fan speed to 95% it lowers gpu engine for 50 and measure mhahseh if drops if doesnt drop than lower again after 15 seconds for 50 again
and measeure mehashes when it drops starts to measure fans speed it lowers for 20 gpu engine now and when reaches fan speed 95% than stops to lower gpu engine if fan speed drops to 90% than rise gpu engine for 20.
The same it is with temperature of GPU only it is much slower... it measure gpu temp and mhases dropping.

[induktor]

Hello, I post this because I'm pretty sure i am doing something wrong.

I cannot control voltage with cgminer in linux.

Linux is xubuntu 12.04, from persistent usb flashdrive, fglrx downloaded from repository, cgminer is Kano's build 3.1.1a.

GPU clock, mem and powertune is OK, but not voltage, whatever I do, it sticks at 1.250V and as a result of this, I am using 100W more of power

My rig is:
3x Sapphire 7950 OC Boost 3G (they are not voltage locked, as I can change it under windows with trixx perfectly in all three cards)
Mother gigabyte 970A-D3, 2 GPUs with 16x risers, 1 with 1x to 16x riser, all powered ones.

I can mine perfectly with the three cards, and control GPU and MEM clock, just cannot change voltage

any ideas, link, tip, idea of where to look for?.
thanks!

[philips]

Hello, I post this because I'm pretty sure i am doing something wrong, no idea what.

I have been using cgminer 3.1.1 under windows perfectly.

Now moved to linux (xubuntu 12.04, from persistent usb flashdrive), same version (Kano's build 3.1.1a) but now I cannot control Voltage.
GPU clock, mem and powertune is OK, but not voltage, whatever I do, it sticks at 1.250V and as a result of this, I am using 100W more of power

My rig is:
3x Sapphire 7950 OC Boost 3G (they are not voltage locked, as I can change it under windows with trixx perfectly in all three cards)
Mother gigabyte 970A-D3, 2 GPUs with 16x risers, 1 with 1x to 16x riser, all powered ones.

fglrx downloaded from repository (last version)

I can mine perfectly with the three cards, and control GPU and MEM clock, just cannot change voltage

any ideas, link, tip, idea of where to look for?.
thanks!

Doesn't work (altering the voltage on Linux) for 6xxx and 7xxx cards. Only solution is a BIOS flash.

[vapourminer]

had this error on my jally a couple times now, and it never recovers on its own. just exiting cgminer and restarting fixes it though. once on one computer (win7 64), then I moved it to another computer (vista 32) and it happened there too.

Code:

[2013-05-25 17:53:51] Accepted a8077b09 Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted bcc62777 Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted 4dcbaa0c Diff 3/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted e350a19d Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:56] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:53:57] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:53:58] BAJ0: Error: Get temp return invalid/timed out (0:-7)
 [2013-05-25 17:53:59] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:00] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:00] BAJ0: GetResults failed (err=-7 amt16)
 [2013-05-25 17:54:01] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:02] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:02] BAJ0: GetResults failed (err=-7 amt16)
 [2013-05-25 17:54:03] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:04] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:04] BAJ0: GetResults failed (err=-7 amt16)
 [2013-05-25 17:54:05] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:06] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:06] BAJ0: GetResults failed (err=-7 amt16)
 [2013-05-25 17:54:07] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:08] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:08] BAJ0: GetResults failed (err=-7 amt16)
 [2013-05-25 17:54:09] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:10] BAJ0: Error: Get temp return invalid/timed out (0:-7)
 [2013-05-25 17:54:11] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:12] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:14] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:15] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:16] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:17] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:18] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:19] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:20] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:21] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:22] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:23] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:24] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:25] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:26] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:27] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:28] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:29] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:30] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:30] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:31] BAJ0: incorrect result count 2 (COUNTemp1: 35, Temp2: 430x00) in (INPROCESS:00x0aCOUNTemp1: 35, Temp2: 430x0aINPROCESS:00x0aCOUNT:00x0aOK0x0a0x00) will try anyway
 [2013-05-25 17:54:31] BAJ0: incorrect result count 35 (INPROCESS:00x00) will try 3 (INPROCESS:00x0aCOUNTemp1: 35, Temp2: 430x0aINPROCESS:00x0aCOUNT:00x0aOK0x0a0x00)
 [2013-05-25 17:54:31] BAJ0: work returned too small (1,INPROCESS:00x00)
 [2013-05-25 17:54:31] BAJ0: work returned too small (1,COUNTemp1: 35, Temp2: 430x00)
 [2013-05-25 17:54:31] BAJ0: work returned too small (1,INPROCESS:00x00)
 [2013-05-25 17:54:32] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:33] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:34] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:34] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:35] BAJ0: incorrect result count 2 (COUNTemp1: 35, Temp2: 420x00) in (INPROCESS:00x0aCOUNTemp1: 35, Temp2: 420x0aINPROCESS:00x0aCOUNT:00x0aOK0x0a0x00) will try anyway
 [2013-05-25 17:54:35] BAJ0: incorrect result count 35 (INPROCESS:00x00) will try 3 (INPROCESS:00x0aCOUNTemp1: 35, Temp2: 420x0aINPROCESS:00x0aCOUNT:00x0aOK0x0a0x00)
 [2013-05-25 17:54:35] BAJ0: work returned too small (1,INPROCESS:00x00)
 [2013-05-25 17:54:35] BAJ0: work returned too small (1,COUNTemp1: 35, Temp2: 420x00)
 [2013-05-25 17:54:35] BAJ0: work returned too small (1,INPROCESS:00x00)
 [2013-05-25 17:54:36] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:38] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:39] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:40] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:41] BAJ0: Error: Request temp invalid/timed out (0:-7)
 [2013-05-25 17:54:41] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:42] BAJ0: RequestResults failed (err=-7 amt0)
 [2013-05-25 17:54:44] MTX: BitForceSC USB failed to release 'cgminer-usb-7-1' err (288)
 [2013-05-25 17:54:44] Shutdown signal received.
 [2013-05-25 17:54:44] 
Summary of runtime statistics:

 [2013-05-25 17:54:44] Started at [2013-05-25 16:31:16]
 [2013-05-25 17:54:44] Runtime: 1 hrs : 23 mins : 28 secs
 [2013-05-25 17:54:44] Average hashrate: 5437.7 Megahash/s
 [2013-05-25 17:54:45] Solved blocks: 0

any ideas?

on the chance is something overheating in the unit Ive put it in the basement with a 140 mm fan on it. the temps were 44-46 C upstairs in the living room (72 F ambient), and 34 C in the basement (65 F ambient with the 140 mm fan). if that fixes it Ill let ya know.

[hahahafr]

Since from 3.1.1 the ability to mine using stratum through a proxy has been lost, and it was my only way to mine, I can't upgrade. What solutions do I have apart from always staying on 3.1.0?

[induktor]

Now moved to linux (xubuntu 12.04, from persistent usb flashdrive), same version (Kano's build 3.1.1a) but now I cannot control Voltage.

Doesn't work (altering the voltage on Linux) for 6xxx and 7xxx cards. Only solution is a BIOS flash.

oops , any ideas where can I found info about how to do this? (flash the card to lower voltage), so I don't bug this thread?
thanks!

[lemonys]

I did a complete reformat and reinstall of my rig with Win XP.  I installed AMD 12.8 (with the bundled OpenCL SDK), but now I have the 100% CPU usage issue.  I understand this is a driver issue, but my question is which driver should I mess with, the AMD video driver or the OpenCL SDK driver?

Thanks in advance.

[philips]

I did a complete reformat and reinstall of my rig with Win XP.  I installed AMD 12.8 (with the bundled OpenCL SDK), but now I have the 100% CPU usage issue.  I understand this is a driver issue, but my question is which driver should I mess with, the AMD video driver or the OpenCL SDK driver?
Thanks in advance.

Tried this?

Q: The CPU usage is high.
A: The ATI drivers after 11.6 have a bug that makes them consume 100% of one
CPU core unnecessarily so downgrade to 11.6. Binding cgminer to one CPU core on
windows can minimise it to 100% (instead of more than one core). Driver version
11.11 on linux and 11.12 on windows appear to have fixed this issue. Note that
later drivers may have an apparent return of high CPU usage. Try
'export GPU_USE_SYNC_OBJECTS=1' on Linux before starting cgminer. You can also
set this variable in windows via a batch file or on the command line before
starting cgminer with 'setx GPU_USE_SYNC_OBJECTS 1'

[stellarman]

I just got my mining rig up and running tonight. I pointed it a P2Pool to mine LTC. I'm getting a decent hash rate (~500kH/s x 2 7950 cards), but I am seeing a lot of "Stratum from pool 0 requested work restart" in the log -- I mean like half of the entries. I'm also getting about a 30% reject rate.

I have tried reducing intensity but that didn't seem to change much except the hash rate. But I haven't tried yet to tweak any of the other settings. The settings I am using are from another guy that is also running 7950s on scrypt to mine LTC. I could post all the setting here if it would help.

First of all, should I be worried? I don't know enough about GPU mining in general, and stratum in particular, to know if what I am getting is normal.

Second, if it is not normal, can anyone tell me what I need to do to fix it?

Thanks in advance for any help you can give.

BTW, I did try to search this thread for that "requested work restart" phrase, but didn't find much that was illuminating.

I thought I would post an update on this, in case anyone else has the same issue.

I made two change, one of which, or the combination of both, solved the problem. The instructions I was using said to set "failover-only" to true. I changed this to "load-balance" : true. At the same time, I added two alternative/backup pools at WeMineLTC.com. When I restarted CGMiner, the "requested work restarts and rejects dropped almost to zero. My rate of accepted shares nearly tripled. It is interesting to note that CGMiner's load-balancing function is now almost completely ignoring P2Pool in the "rotation" of pools. I am now at ~600kH/s per GPU with single digit rejects.

If anyone else has a similar problem, those are two things to try.

[philips]

...any ideas where can I found info about how to do this?

Jump in here : https://bitcointalk.org/index.php?topic=205406.msg2237983#msg2237983

[lemonys]

Tried this?

Q: The CPU usage is high.
A: The ATI drivers after 11.6 have a bug that makes them consume 100% of one
CPU core unnecessarily so downgrade to 11.6. Binding cgminer to one CPU core on
windows can minimise it to 100% (instead of more than one core). Driver version
11.11 on linux and 11.12 on windows appear to have fixed this issue. Note that
later drivers may have an apparent return of high CPU usage. Try
'export GPU_USE_SYNC_OBJECTS=1' on Linux before starting cgminer. You can also
set this variable in windows via a batch file or on the command line before
starting cgminer with 'setx GPU_USE_SYNC_OBJECTS 1'

I forgot to mention - yes I did try that before I made the post.  It had no effect.  The rig was running on Catalyst 12.8 before I reformatted it, although I don't remember the SDK version I was using.  I'll start playing with the SDK version and see if that fixes it.

[kano]

had this error on my jally a couple times now, and it never recovers on its own. just exiting cgminer and restarting fixes it though. once on one computer (win7 64), then I moved it to another computer (vista 32) and it happened there too.

Code:

[2013-05-25 17:53:51] Accepted a8077b09 Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted bcc62777 Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted 4dcbaa0c Diff 3/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted e350a19d Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:56] BAJ0: RequestResults failed (err=-7 amt0)
...

any ideas?

on the chance is something overheating in the unit Ive put it in the basement with a 140 mm fan on it. the temps were 44-46 C upstairs in the living room (72 F ambient), and 34 C in the basement (65 F ambient with the 140 mm fan). if that fixes it Ill let ya know.

I've increased the timeout value (-7 = timeout) for windows for all devices that weren't already 999ms since it seems windows and USB sux badly
(Linux is 100ms or 200ms and works fine even on an rpi)

Other possible causes could also be:
USB HDD/large bandwidth device blocking/delaying the USB bus with large transfers
Computer bogged down, not much CPU available
Computer slowed down due to not enough RAM, swapping
Plugged into a USB1.1 port (not sure how bad USB1.1 is on windows)
Plugged into a USB3.0 port (problems as yet not resolved)

Anyway, it will be 999ms in the next release for all windows devices
Yes that means wait for a whole second for replies from the device (if there is a problem)
... 999ms is more than a full nonce range on a Jalapeno ...

Edit: and before anyone assumes something incorrect, this problem is the same for both USB and the old serial-USB, makes no difference

[emunebtk]

I have searched far and wide and have not found and answer to solving my 40 Kh/s on a ATI Radeon 5770 problem via CGminer 3.1.1 on osx 10.7 Lion Mac Pro 1,1 .

This is my script line:

Code:

export GPU_MAX_ALLOC_PERCENT=100 
export GPU_USE_SYNC_OBJECTS=1 
cgminer --scrypt -w 64 --lookup-gap 2 -g 1 --thread-concurrency 3200 -I 12 -o http://litecoinpool.com:9332 -u worker.1 -p xxxxxx

I keep reading about adding : --auto-gpu --gpu-engine --gpu-memclock --auto-fan --gpu-fan  --temp-target --temp-overheat --temp-cutoff ,etc.

and can not get any of those to work with CGminer. Upon adding any of the above Cgminer will not open, and adding anything above Intensity 12 crashes the computer? Any ideas? Would love to reach 160

[vapourminer]

had this error on my jally a couple times now, and it never recovers on its own. just exiting cgminer and restarting fixes it though. once on one computer (win7 64), then I moved it to another computer (vista 32) and it happened there too.

Code:

[2013-05-25 17:53:51] Accepted a8077b09 Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted bcc62777 Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted 4dcbaa0c Diff 3/1 BAJ 0 pool 0 
 [2013-05-25 17:53:55] Accepted e350a19d Diff 1/1 BAJ 0 pool 0 
 [2013-05-25 17:53:56] BAJ0: RequestResults failed (err=-7 amt0)
...

any ideas?

on the chance is something overheating in the unit Ive put it in the basement with a 140 mm fan on it. the temps were 44-46 C upstairs in the living room (72 F ambient), and 34 C in the basement (65 F ambient with the 140 mm fan). if that fixes it Ill let ya know.

I've increased the timeout value (-7 = timeout) for windows for all devices that weren't already 999ms since it seems windows and USB sux badly
(Linux is 100ms or 200ms and works fine even on an rpi)

Other possible causes could also be:
USB HDD/large bandwidth device blocking/delaying the USB bus with large transfers
Computer bogged down, not much CPU available
Computer slowed down due to not enough RAM, swapping
Plugged into a USB1.1 port (not sure how bad USB1.1 is on windows)
Plugged into a USB3.0 port (problems as yet not resolved)

Anyway, it will be 999ms in the next release for all windows devices
Yes that means wait for a whole second for replies from the device (if there is a problem)
... 999ms is more than a full nonce range on a Jalapeno ...

Edit: and before anyone assumes something incorrect, this problem is the same for both USB and the old serial-USB, makes no difference

could be a couple of those. the win7 desktop might of been a USB 3 port (didnt notice, it was plugged into the back I/O which has both) and the vista HTPC box was under a decent load (recording and streaming, but nothing on USB though),

since being banished to the basement, to an ancient Pentium 4 box with a BFL FPGA single, 5830 and now the jally, its been fine. maybe being cooler helps too, dunno. it will stay down there now, its a bit too noisy for my tastes.

just curious, you seem to read 2 temps, and theres a 7 degree C difference between them. those are the two hashing chips? if so Ill have to tear it down and do the TIM  right.

[kano]

...
since being banished to the basement, to an ancient Pentium 4 box with a BFL FPGA single, 5830 and now the jally, its been fine. maybe being cooler helps too, dunno. it will stay down there now, its a bit too noisy for my tastes.

just curious, you seem to read 2 temps, and theres a 7 degree C difference between them. those are the two hashing chips? if so Ill have to tear it down and do the TIM  right.

Yeah I hear (but have not seen) that temp can be a problem but it's heading to winter shortly here in Aus, so I've not really done any bad temp testing on it.

The amt0 - which now says amt=0 in the next version correctly - means it literally got no reply at all, not some unexpected reply.
It looks like after that it got out of sync for 13 seconds then seemed to solve the temp out of sync but get no work replies.
Lastly it's getting Temp and Work replies mixed together so certainly something went amiss.

The MTX: problem is the last thing I need to fix before I'll consider the USB code ready for another version release, then it'll be up to ckolivas to decide what and when

Edit: yeah on linux it runs like a dream

[Paladin69]

Hi,

I haven't upgraded since 2.10.4  The latest version is very different.  fpgaonly.exe is now replaced by nogpus.exe, I guess?  The way you start the COM's must be different or something?  Getting errors and nothing wants to start.

I used to type this...

cgminer-fpgaonly.exe -o http://www.xxx.com:port -u user_worker -p pass -S bitforce:\\.\COM3 -S bitforce:\\.\COM4 -S bitforce:\\.\COM5 -S bitforce:\\.\COM6 -S bitforce:\\.\COM7 -S bitforce:\\.\COM8 -S bitforce:\\.\COM9 -S bitforce:\\.\COM10 -S bitforce:\\.\COM11 -S bitforce:\\.\COM12 -S bitforce:\\.\COM13 -S bitforce:\\.\COM14 -S bitforce:\\.\COM15 -S bitforce:\\.\COM16 -S bitforce:\\.\COM17 -S bitforce:\\.\COM18 -S bitforce:\\.\COM19 -S bitforce:\\.\COM20 -S bitforce:\\.\COM21 -S noauto

Doesn't work now.

Sorry the help files are just a sea of words to me.  Can somebody look at the string above and tell me what to modify?  This is for Windows.

[zvs]

Is the mother replacement for BAMT anywhere other than http://ck.kolivas.org/apps/cgminer?  I'm not sure if it's just been down for a few hrs, days, or what.  

oops , any ideas where can I found info about how to do this? (flash the card to lower voltage), so I don't bug this thread?
thanks!

Radeonvolt doesn't work?  Seems like it should, if a BIOS flash does

[kano]

Hi,

I haven't upgraded since 2.10.4  The latest version is very different.  fpgaonly.exe is now replaced by nogpus.exe, I guess?  The way you start the COM's must be different or something?  Getting errors and nothing wants to start.

I used to type this...

cgminer-fpgaonly.exe -o http://www.xxx.com:port -u user_worker -p pass -S bitforce:\\.\COM3 -S bitforce:\\.\COM4 -S bitforce:\\.\COM5 -S bitforce:\\.\COM6 -S bitforce:\\.\COM7 -S bitforce:\\.\COM8 -S bitforce:\\.\COM9 -S bitforce:\\.\COM10 -S bitforce:\\.\COM11 -S bitforce:\\.\COM12 -S bitforce:\\.\COM13 -S bitforce:\\.\COM14 -S bitforce:\\.\COM15 -S bitforce:\\.\COM16 -S bitforce:\\.\COM17 -S bitforce:\\.\COM18 -S bitforce:\\.\COM19 -S bitforce:\\.\COM20 -S bitforce:\\.\COM21 -S noauto

Sorry the help files are just a sea of words to me.  Can somebody look at the string above and tell me what to modify?  This is for Windows.

With 3.1.1 Switch the driver with zadig and remove every -S except the last one (-S noauto)
... read README, FPGA-README

[vapourminer]

just curious, you seem to read 2 temps, and theres a 7 degree C difference between them. those are the two hashing chips? if so Ill have to tear it down and do the TIM  right.

Yeah I hear (but have not seen) that temp can be a problem but it's heading to winter shortly here in Aus, so I've not really done any bad temp testing on it.

The amt0 - which now says amt=0 in the next version correctly - means it literally got no reply at all, not some unexpected reply.
It looks like after that it got out of sync for 13 seconds then seemed to solve the temp out of sync but get no work replies.
Lastly it's getting Temp and Work replies mixed together so certainly something went amiss.

The MTX: problem is the last thing I need to fix before I'll consider the USB code ready for another version release, then it'll be up to ckolivas to decide what and when

Edit: yeah on linux it runs like a dream

yeah yeah linux, I know..  us poor windows slobs

Ive played with linux (netbook is dual boot) but I have a few old XP boxes, that P4 being one of them. its funny, built it in 2005 (?) and its still going strong. cant remember the last time I rebooted it, and the only time cgminer is shut down is to install new versions. I use your API to keep tabs on it.

anyway, no worries. its going fine now. when the jally was upstairs the case was rather warm to the touch, its room temp now. typical crappy BFL cooling: looks over performance. I will probably order slim 92mm high flo ball bearing fan for it.