Hi everyone,
A new miniZ version v1.9z2 is out with performance improvements for AMD GPUs, on Flux!
Fixed major bug causing miner to fail start, on Windows.
Please find miniZ version v1.9z2 @ Download page.
Changelog:
- Performance improvements for some AMD gpus, on Flux. Major improvements for the RX 6800 XT.
- Fixed issues that were causing the miner to fail start, on Windows.
- Fixed issue interfering with Zano mining.
- Removed ‘automatic’ clock reset, on exit. Except when they are changed on the command line.
*** Thank you all for the feedback! ***
Remember to try --oc1/--oc2/--ocX options for optimum performance. (NVIDIA only)
For additional information check our Usage or FAQ pages.
*** Download miniZ latest version here. ***
Engage, feedback and support
Because your feedback is very important, miniZ kindly encourages you to use the comment box below for general questions and feedback.
Certainly we will strive to help you with any doubt and provide a solution to any problem you may encounter. In case you need to reach us, please check the Support page.
Follow us
On Twitter
Our thread on bitcointalk.org
We wish you a fast and friendly mining experience!
Can you make a command line parameter to hide clock and memory speeds? This is not needed on every line. It clutters up the display. Thanks!
Hi David Szczesny,
We’ll implement this for the next version.
Thanks for the suggestion.
Cheers
I have found a lot of instability on the 2000 series Nvidia cards, especially the 2070’s Always hard crashes the miner after a couple hours, can’t pinpoint the problem, seems like almost every one of my 2070’s are not stable
Zane, is your instability on the new v19z2 version? I’m still on v19z and I’m also having issues with my 8x 2060 rig. It crashes ever few hours with conservative overclocks. My 2080ti/30xx rig is rock stable though. Waiting for Hive to push the miner update.
No I have instability on earlier versions as well, I’ve also tried with conservative or no overclocks, still crashes.My 2080Ti and 1070 are also rock stable, but most of my 2070 have problems no matter what I try, Risers, different version, stock OC.I am using windows 10, mining Flux, I just end up pointing my 2070’s at eth(runs stable) until I can figure out what is causing the problem, I should note other mining software seems to be stable but I haven’t tried a long duration test on other miners yet.
Hi zane,
We’ll test this and try to understand what is happening.
Thanks for the information.
Cheers
Hi Dizzle,
Let us know if this persists with the v1.9z2.
Maybe check if this could be OCs, or even the power supply working on the edge.
In any case we’ll have a look. Are you mining flux?
Cheers
miniZ,
Thanks for the automatic clock reset fix! I just upgraded one of my rigs(3x 2080ti,4x 3060, 1x3060ti) to v1.9z2. I only experienced one crash on this rig, mostly stable. However, my other rig(8x 2060, 1x 3060ti) is very unstable on v1.9. I tried underclocking, overclocking and stock clocks, the rig would hard crash and hang Hive 2-3 times a day. Last night I rolled that rig back to v1.8z3 and its rock stable, no crashes. I’m mining BTG zhash 144,5.
update, 1.8z3 is also unstable on 144,5 on my 2060 rig. i swapped out my hive usb with a hive ssd and it still crashes. not sure what the issue is.
Hi Dizzle,
Thanks for the information.
Do you get any error? If so, could you paste it here?
We’ll have a look.
Cheers
edit: Maybe you can try to lower the gpu power limit a bit.
Any chance to add CuckooCycle for Cortex (CTXC)?
Hi Decidueye,
At the moment we will not add, but we’ll think about it.
Thanks for the suggestion!
Cheers
PLEASE ! Why miniz does not have commande lines to set the clocks like GMINER ??? With Miniz we have to set everything in Afterburner when the RIG is on Windows, it is dangerous because if afterburner fail all the power limits goes back to 100% and may cause severe damage.
They actually do have locked core and power limit settings in miniz, the issue is they are missing core offset settings. Without that the others are kinda useless.
By setting the core offset to 200+ you can gain 5-10 sols on most cards I have tested at the same power limit, thus greatly improving efficiency.
Until this is added I am stuck using Trex on ergo since I must have a 1 click setup on my gaming machines.
Hi Alex,
You can set PL with miniZ too. Use
--power, or-pl, option.We’ll see what we can do about this.
Thanks.
Cheers
Hi Alex, Overclocker,
New version in HiveOS/Ubuntu still showing my 6700xt with a 0% fan speed in the miner. Fan is definitely working Skipping miner log rotation due to execution time < 30secNumber of miniZ CUDA>=[8.0] using driver CUDA[11.60] devices found: 5 #0 Nvidia GeForce RTX 3070 Ti | SM version: 8.6 | SM count: 48 | cores: 6144 | Memory: 7.630GB | BusID: 0000:01:00.0 #1 Nvidia GeForce RTX 3060 Ti | SM version: 8.6 | SM count: 38 | cores: 4864 | Memory: 7.657GB | BusID: 0000:08:00.0 #2 MSI GeForce RTX 3060 Ti | SM version: 8.6 | SM count: 38 | cores: 4864 | Memory: 7.657GB | BusID: 0000:0a:00.0 #3 Nvidia GeForce RTX 3060 Ti | SM version: 8.6 | SM count: 38 | cores: 4864 | Memory: 7.657GB | BusID: 0000:0b:00.0 #4 EVGA GeForce RTX 3070 FTW3 Ultra | SM version: 8.6 | SM count: 46 | cores: 5888 | Memory: 7.636GB | BusID: 0000:0c:00.0 #5 AMD Radeon RX 6700 XT | gfx1031 | SM count: 20 | cores: 0 | Memory: 11.984GB | BusID: 0000:07:00.0Number of OpenCL[Advanced Micro Devices, Inc.] devices found: 1miniz exited (exitcode=0), waiting to cooldown a bit
Hi David Szczesny,
We also answered to your message in the other post.
Thanks for the information.
We’re investigating.
Cheers
Thanks! I’ve seen anther Discord user that have the same issue on an nvidia 8 card rig. It might not be isolated to just AMD – but however does affect my only AMD card.
Using MiniZ I seem to lose several mh on Ethash and Equihash algos using CMP 170HX cards. Can the miner be a bit more optimized for these? Also, is it possible to add ubqhash since it is basically Ethash? Thanks a lot
Hi Tswift,
Which version are you using?
Could you paste here the output of running:
./miniZ.exe -ciWe’ll see what we can do about optimizing for the CMP 170X card, and regarding Ubiq.
Cheers
Can you add support for memory temps to be visible in hiveos?
Hi Archer,
Thank you for your suggestion.
For the next version, we’ll not be able to implement this. But, we’ll see what we can do for future releases.
Cheers
Hi, please upload to the Github. very important for Awesome Miner users. Thank you.
Hi Alex,
Done!
Cheers
Previously I used version 1.8z3/linux/RaveOS/Equihash 144.5/nvidia gtx1070. After updating to version 1.9z2, the miner writes performance 11-12 times higher, consumption is two times lower. But not a one share was found in an hour, although the pool of diff is such that every minute must be determined. Then I installed version 1.9. She has a problem on it. I had to go back to 1.8z3. Please correct the mistakes. Thank you.
Hi Serg,
Thank you for the information. Are you still experiencing this issue?
So far, we could not replicate this behavior.
Which pool are you using? Could you also paste here your command line?
We’ll keep looking into it.
Cheers
Please add zil split mining ability with flexpool’s new ZMP stratum. I would like to mine flux/zil with miniZ
Hi Mike,
We’re not sure we’ll implement this at the moment, but we’ll have a look.
Thanks for the suggestion!
Cheers
Why on hiveos, if i select miniZ as my miner,
kawpow does not show up in the drop down menu of the algorithms?Hi Mo,
Thank you for the information.
This is a question for HiveOS. We’ll see what we can do about this.
Cheers
Hello! I am switching from ETH today, and I just downloaded this newest version to try mining Flux. However, when I launch the .bat file, the miner doesn’t even properly start before it freezes my entire computer, forcing a hard restart. I am using Windows 10, and only NVIDIA cards. I tried it both with my cards OC’d before launch, and at default settings on launch. I have just the basic settings for the pool I’m mining to in my .batAm I doing something wrong? Is there a specific command I need to run to stop this from happening? I was previously using PheonixMiner to mine ETH, and never experienced this with my cards.
Hi Sonavyx,
Thank you for your feedback.
We’ll have a look.
Could you paste here your command line?
Maybe also the output of running:
miniZ.exe -ciCheers
ps: OCs for Equihash are usually different from Ethash algorithm. Resources required for each algorithm vary a lot.
With the new fan control feature, it requires an updated driver version.
Is it possible to allow fan control for older drivers? I’m running 512.15 and have found this to be a stable driver.
Hi Mike,
We may be able to do this for Windows.
We’ll have a look.
Cheers
Thank you, im looking forward to it.
I am running release v1.9z2 and it is crashing running on my Nvidia 2080TI FE with one of the following errors: [INFO ] Mining fee set to 2.00%miniZ<125,4>[06:0:00.0:10841]: Selecting GPU#0[0] Nvidia GeForce RTX 2080 Ti[FATAL ] GPU[0]: CUDA error 73 ‘an illegal instruction was encountered’ in line 296[INFO ] QUIT miniZ[INFO ] Disconnecting from us-flux.fluxpools.net[FATAL ] GPU[5]: CUDA error 77 ‘an illegal memory access was encountered’ in line 296[INFO ] QUIT miniZ[FATAL ] GPU[2]: CUDA error 77 ‘an illegal memory access was encountered’ in line 296[INFO ] QUIT miniZ/hive/miners/miniz/h-run.sh: line 20: 12730 Segmentation fault (core dumped) ./miniZ $(< {MINER_NAME}.conf) –logfile $MINER_LOG_BASENAME.log –telemetry 127.0.0.1:${MINER_API_PORT} –gpu-line
Hi Wayland,
These errors (CUDA error) are usually OC related.
Try to adjust the OCs a little, and check again.
But could you paste here the output of running
miniZ.exe -ciThanks.
Cheers
ps: OCs for Equihash are usually different from Ethash algorithm. Resources required for each algorithm vary a lot.
root@BMO:/hive/miners/miniz/1.9z2# ./miniZ -ciNumber of miniZ CUDA>=[8.0] using driver CUDA[11.50] devices found: 1 #0 Nvidia GeForce RTX 2080 Ti | SM version: 7.5 | SM count: 68 | cores: 4352 | Memory: 6.688GB | BusID: 0000:01:00.0
I was using with a little modification the OC setting that work on my 3090 and miniZ mining flux on HiveOS. I changed the settings to Core 180, Mem 2200, PL 85 and it has been running stable now for 10 hours ad about 96 Sol. On the 3090 i get a temprature reading for both Core and Mem, however on the 2080ti FE I only receive a temperature reading for the Core. Is that a miniZ issue or a HiveOS issue?
Hi Wayland,
Thanks for the feedback. Is all running well for you still?
miniZ does not show mem temperature. This should be HiveOS.
We’re not sure but we think that mem temperature will only show for the RTX 30XX.
Cheers
The rig is stable once I adjusted the OC settings.Thanks for your followup.
I spoke to soon, it was running stable for about a week and now I have received this error three times today. [FATAL ] GPU[0]: Driver error 999 ‘unknown error’ in line 178 [INFO ] QUIT miniZ
Hi Wayland,
This error is related o OCs.
Have you tried to adjust them a bit more?
Also, have you tried the new version v1.9z3
Cheers
Any plans to support RX 6600/XT? This miner is awesome on 6800 XT! excellent efficiency.
Hi CJ,
Thank you for your message.
miniZ already supports RX6600/XT. Are you having any error message?
Could you give some more information on what is happening? Which algo, and OS? Maybe also which pool, depends on the error.
Cheers
hello, I am unable to use –fanspeed=100 -pl 70 in windows, I get the messageAnd I have rune as an admin…************ miniZ v1.9z2 ************Number of miniZ CUDA>=[8.0] using driver CUDA[11.60] devices found: 2Number of OpenCL[] devices found: 0[ERROR ] GPU[00] cannot adjust power![ERROR ] GPU[00] cannot adjust fan, this driver version does not allow changing fan speed!
Hi Trippy,
We may be able to make this work for older drivers, on Windows.
We’ll have a look.
Cheers
Hello, I just migrated my rigs under Windows on Flux with miniZ. I set up the conf file and telemetry.
However, I am unable to include the arguments either in the bat file or the conf file :
–fanspeed=[fanspeed(%)]
–gpuclock=[gpuclock(MHz)]
–memclock=[memclock(MHz)]
–power=[power(W)]
-pl [power(W)]
Thank you for your help.
Hi Pika28,
At the moment these options do not work with config file, but we’ll have a look. However, you can use them with bat file.
You need to run the file with administrator privileges.
Thanks for the feedback.
Cheers
Hello, I just downloaded this version of MiniZ miner and wanted to give it a try but for some reason miner does not apply the memory clock that I need. I’m on Windows 10 and I’m testing it on my EVGA 3060 Ti. I tried setting the clock relatively memclock=1000 and memclock=+1000 but that just locks my memory clock to 5000 MHz and I have to reboot to reset it. I also tried an absolute value memclock=8000, but despite what I set I’m getting 6800 MHz clock.Can you please help my understand what I’m doing wrong?
Hi John,
Thank you for your message.
Sorry about this, we’re looking into it.
--memclockoption needs absolute values, but is not working well.In the next version we will add an option to set the offset (not absolute values) on Windows.
Setting it to ‘0’ (zero) will restore the default value.
We expect the new version to be out by the end of next week.
Cheers
I’m having the same problem. I cannot set any memory clocks in miniZ at this time. Makes it impossible to use.
Hi draven,
Could you try the new options (version v1.9z3)
--memoffsetand--gpuoffset? Maybe these are more suitable for what you are looking for.Thanks for the feedback.
Cheers
Hello there. The miner does not set clock values, power limit and fan speeds to default values on Windows 10 after exiting/crashing. Is there a way to fix this?
Hi Jack,
Power and fan speed do not reset on exit. You’ll need to specify the values you want in the next run.
Could you try the new options
--memoffsetand--gpuoffset? Maybe these are more suitable for what you are looking for.Because this is an ofset, if you set them to 0 it will restore the default stock values.
Thanks for the feedback.
Cheers
Hello, I just downloaded this version of MiniZ miner and wanted to give it a try but for some reason miner does not apply the memory clock that I need. I’m on Windows 10 and I’m testing it on my EVGA 3060 Ti. I tried setting the clock relatively memclock=1000 and memclock=+1000 but that just locks my memory clock to 5000 MHz and I have to reboot to reset it. I also tried an absolute value memclock=8000, but despite what I set I’m getting 6800 MHz clock. Can you please help my understand what I’m doing wrong?
Hi John,
Could you try the new options (version v1.9z3)
--memoffsetand--gpuoffset? Maybe these are more suitable for what you are looking for.Thanks for the feedback.
Cheers
I tried miniZ miner with ETHW mining instead of T-rex like I normally do and I noticed that all of my hashrates were much lower than T-rex using the same overclocks. Sometimes 30-40% lower hashrate. All of my cards are nvidia non-LHR, and I use locked core clock with them all. Is the locked core a problem with miniZ? I noticed other miners like lolminer, gminer, nbminer don’t seem to have this issue.
Hi Tswift,
Could you try adding
--tune=0to your command line and let s know if this helps?Thanks!
Cheers
Any chance of showing the highest share difficulty and found by GPU to the next release?
Hi Dizzle,
Thank you or the sugestion.
We do not find it particularly useful.
Could you share why would you need this information?
Thanks!
Cheers
I finally upgraded to miniZ 2.0 and man is that great! It makes my 5700s hash well above my 10 series and almost on par with my 20 series cards without any changes to my OC. Excellent work!
Hi Andy Beezy,
Thanks for the feedback.
Great that it’s working for you.
Cheers