That's my man, i really love it when someone takes time and tries himself something
So it looks like the binary kernel is faulty ? I will check that out.

Thank you for your efforts

edit:
And YES, the binary IS faulty. Great that i pushed the 1.6.8 version 10 days before the fork so we can smooth these kind of things out

so what is the conclusion at the end?
1.6.8 hashing better on rx550 and rx560 cards ?

also where did you get intensity 30 from?
You are really confusing me here i don;t know even what algo are we talking about now.

550 4g should be ok on V7 with : 27/12/2
560 4g should be ok on V7 with : 27/8/2

550 4g should be ok on V8 with : 28/16/2
560 4g should be ok on V8 with : 29/32/2

Try maybe worksize 32. I don't have vega64 to test with, but i have a 56, and the best hash for it is at 120/16, it should be around the same for 64 too.

Intensity 54 too much with 2 threads. Why not 27 ?

have you tried auto intensity for 550?

also it would be nice if you could write your settings

Yes something not ok, try setting by hand for example 120/16 2 threads

I really hurried to push the v8 version, so of course bugs are to be expected, and we still have time to fix them until the fork

Ok going to re-check and test.

Can you test on bittube too?

interesting..and on 1.6.7 you get the 1000+ speed on all cards ?

Wait 10-15 min for it to get to the max speed.
Nothing has changed for heavy algos in the code.

ill have to lower that then , 30 is good for 4gb

Have you read the big red text on the first page that say which driver versions NOT to use when compiling?

V1.6.8
- New Monero algo added (normalv8)
- Small speed increase on some cards
- Fee is now ~0.85% for both normal mode and algo switching mode
- Faster miner startup
- Pools with different algos can now be used in pools.txt, miner will re-init on switch
- Added a new parameter for gpu tuning in gpu_conf : bralock
- Rewritten release of resources on miner shutdown/restart
- Added parameter --forcedshutdown if you don't want to release resources on exit
- Fixed bug when setting compute mode on more than 9 cards
- Fixed a bug showing bad date for 'connected since' and 'average share find time'
- Gpu hashrate watchdog now won't trigger if user is not connected to a pool
- reboot_script parameter removed and added 'reboot_script_min_rig_speed' and 'reboot_script_gpu_watchdog' instead
- Added more logging in gpu init part
- API parameters now have to be set in start.bat (--apienable, --apiport, --apirigname) , removed them from config.txt
- Press 'o' to switch to the previous pool from pools.txt
- If you add 'stratum+ssl://' before the pool address, TLS will be automatically enabled

+ Monero will fork on block 1685555 (that is approx Oct 18th). A new algo is introduced, use 'normalv8' for cryptonight_type
+ Fee is now equal for both normal and algo switching mode (~0.85%)
+ Miner now starts up a little bit faster
+ Pools can now use cryptonight_type, and if the next/previous pool algo isn't the current one, miner will restart with the correct settings
There is a short explanation on how to use this option on the first page and in the readme.txt

+ Added parameter bralock in gpu_conf, it is a value from 1-256, it *may* give you a little speedup if you find the right value for your algo/gpu
+ Removed the reboot_script parameter, and instead made 2 new parameters :
"reboot_script_gpu_watchdog" : filename to a batch file in miner directory, if set it turns off built in miner reset procedure on gpu failure, and instead runs this script
"reboot_script_min_rig_speed" : filename to a batch file in miner directory, if set it turns off built in miner reset procedure on min_rig_speed trigger, and instead runs this script

+ API parameters removed from config file, so they don't have to be defined in every config separately.
Now you can set them in start.bat with :

--apienable (enable statistics API)
--apiport value (port where statistics API is reachable - default 21555)
--apirigname value (identifier name for your rig in statistics API)

+ Added possibility to switch to the previous pool from your pools list by pressing 'o'. If you get to the first pool on the list, and press 'o' it will go around, and jump to the last pool.

Also i implemented a temporary solution for the transition from V7 to V8 algo, so the miner can switch automatically when the fork happens.
You can read it HERE on how to do it.

Basically you set the config for normalv8 in Config folder(intensity and everything else), enable the algo switching (by adding --usealgoswitching in start.bat), and create a temporary pools file, where you put 2 pools, the first one is the V7 pool, the second one is V8.

Set start.bat to use that temporary pools file, and when the time comes, miner will detect if the received job is for the new algo, and will switch to the next pool on the list, which in this case is the pool using the V8 algo.

Also the config-normalv8 settings will apply.

You should really open the readme.txt that comes with the miner.
Also on the first page you can find some settings for various cards on various algos, also some for Vega too.

As with the power consumption, you need to tweak a few things (properly set overdriventool, find good intensity/power ratio etc..), it won't set everything magically for you, sorry.

No, sorry.

I guess i can add every version from 18.7.1 up to the latest one. Some guys already reported to AMD, but i guess it won't be fixed really soon.
Probably could make it compile on these versions too, but i would have to rewrite probably like everything, and who knows what would the performance be like.

Here is your problem.

A few post before yours :


Claymore works because he is not compiling the kernels, he has prebuilt binaries.

It should work

Hi, thanks for the kind words.
Yes, from drivers 18.7.1 or 18.8.1 the binaries created are bad, the compiler misinterprets something very badly.
But you can use the binaries built on older driver versions, they will work.

The new version will have an autoswitch, it will be a temporary trick only, that will be removed later in newer miner versions, i will add it
just to make the transition from v7 to v8 smooth as possible.

Someone posted another version of the logtime creation, i will use that from next version

Oct 18th