jjxaker Posted February 18, 2017 Posted February 18, 2017 Hello! Bug reading vcore, It displays an incorrect value. I wanted to create a aida64 full report, but in reading SMBus dump(full) my system herself shuts down. but this is clearly a problem of the aida64... isasensordump.txt Quote
Fiery Posted February 18, 2017 Posted February 18, 2017 We follow MSI's recommendations on how to handle the sensor registers on MSI Z270 motherboards. What exactly is the issue about the Vcore? Does MSI's own monitoring software measure a different value or the same value? Quote
jjxaker Posted February 18, 2017 Author Posted February 18, 2017 any other software reads correctly... in the BIOS manually voltage 1.360 PS. lacks +3 V the same problem with the ASUS PRIME Z270-K cpu vid read exactly what is needed! cpu core voltage 1.144 It shows the nonsense. Quote
Fiery Posted February 20, 2017 Posted February 20, 2017 Please use MSI's own software as a reference on MSI motherboards, and Asus' own software as a reference on Asus motherboards. We'll add CPU VID voltage below the CPU Vcore reading to the Sensor page as well, so you can see both values listed. Quote
jjxaker Posted February 20, 2017 Author Posted February 20, 2017 3 hours ago, Fiery said: Please use MSI's own software as a reference on MSI motherboards Software and utility from manufacturers always been crap, how to use them! it is for housewives))) but monitoring is working properly. Quote
Fiery Posted February 21, 2017 Posted February 21, 2017 We've contacted MSI to find out what causes the discrepancies between the information they've provided us, and the way their own monitoring software works. Meanwhile, we've added the missing +3.3V reading in the latest beta version of AIDA64 Extreme available at: https://www.aida64.com/downloads/latesta64xebeta After upgrading to this new version, make sure to restart Windows to finalize the upgrade. Quote
jjxaker Posted February 21, 2017 Author Posted February 21, 2017 Thank you, I hope you and MSI will find the truth with. Temporarily, and so will come down, true looks CPU VID of reading more than vCore Quote
Fiery Posted February 22, 2017 Posted February 22, 2017 It seems on your particular motherboard model the Vcore reading has to be multiplied by 1.2. We'll implement that in the next AIDA64 beta update Quote
jjxaker Posted February 22, 2017 Author Posted February 22, 2017 Need to replace the VCORE on VID. VID is always less than VCORE Quote
Fiery Posted February 23, 2017 Posted February 23, 2017 Since they both will be displayed on the Sensor page, I don't think we need to replace anything Whichever you don't like, just ignore it Quote
jjxaker Posted February 26, 2017 Author Posted February 26, 2017 Thank you! new beta 5.80.4084 solved the problem of monitoring the voltage. I have a question about Cache & Memory Benchmark, incorrectly displays the frequency North Bridge Clock. but this problem has long. Quote
Fiery Posted March 6, 2017 Posted March 6, 2017 In case the north bridge clock fluctuates, it's normal that sometimes AIDA64 cannot capture the highest value in the power states. Or is it consistently lower on the Cache & Memory Benchmark panel than the value displayed on the Overclock page? Quote
jjxaker Posted March 6, 2017 Author Posted March 6, 2017 Yes, north bridge clock he always shows like that. Quote
jjxaker Posted March 13, 2017 Author Posted March 13, 2017 asusspddump_test1.txt asusspddump_test2.txt And how about the problem with NB Clock? This is a problem with MSI only. Quote
Fiery Posted March 14, 2017 Posted March 14, 2017 I'm afraid we have no idea what could work incorrectly about the NB clock measurement. How do you know it's an issue limited to MSI motherboards? As for the SPD issue, for some reason your memory modules cannot perform SPD bank switching. Bank switching is necessary to read the full 512-byte SPD block of DDR4 memory modules. Can you please check if CPU-Z is also affected? Quote
jjxaker Posted March 14, 2017 Author Posted March 14, 2017 4 hours ago, Fiery said: How do you know it's an issue limited to MSI motherboards? There are no problems NB clock with Asus Z170/Z270, in contrast to MSI Z170/Z270 Series... I tried the boards MSI Z170 Gaming M3/M5/M7/M9 and Z270 Gaming M5/M7 with all of them the frequency of the north bridge is read incorrectly. 4 hours ago, Fiery said: As for the SPD issue, for some reason your memory modules cannot perform SPD bank switching. Bank switching is necessary to read the full 512-byte SPD block of DDR4 memory modules. This may be the problem due to bios? I pulled out the video card(EVGA GTX 1070 FTW) and the problem disappeared. At least I can not reproduce the situation with the problem... ps. But is it right that it is indicated that there is not present thermal sensor? Because it is available, Giantec GT34TS04 CPU-Z when there was a problem not always read the data SPD... Now for now everything is fine Quote
Fiery Posted March 14, 2017 Posted March 14, 2017 The difference between how NB clock is measured on the Overclock page vs. on the Cache & Memory Benchmark panel, is that the latter will disable the power management facilities (namely EIST a.k.a. SpeedStep and C1E a.k.a. Enhanced Halt State) of the CPU before measuring the clock speeds. Maybe MSI motherboards alter the behavior of the NB, ie. adjust its operating frequency in a proprietary way when those power management features are disabled? As for the SPD bank switching issue, it looks a lot like a BIOS issue. Quote
jjxaker Posted March 14, 2017 Author Posted March 14, 2017 24 minutes ago, Fiery said: Maybe MSI motherboards alter the behavior of the NB, ie. adjust its operating frequency in a proprietary way when those power management features are disabled? EIST disabling It helps. But this is a bad way out of the situation))) Miracles are what by the way vid and vcore equaled after turning EIST off Quote
Fiery Posted March 14, 2017 Posted March 14, 2017 I suppose the BIOS is doing some magic on your MSI motherboard Quote
jjxaker Posted March 14, 2017 Author Posted March 14, 2017 Unfortunately they communicate with them is difficult... I can not even Make the SMBus dump(Full), the system itself turns off Quote
Fiery Posted March 14, 2017 Posted March 14, 2017 23 minutes ago, jjxaker said: Unfortunately they communicate with them is difficult... I can not even Make the SMBus dump(Full), the system itself turns off If you could check what's the device that's scanned when the system turns off, we can filter that device out. Previous MSI boards with Z170 and X99 chipset are sensitive for scanning at device 21h, 22h and 23h, so we skip those devices there. We haven't received any guidelines from MSI about Z270 boards, so we somehow have to discover if there are sensitive devices there too Quote
jjxaker Posted March 14, 2017 Author Posted March 14, 2017 B00-D1E/Reg 00D1 , And set off a sharp switch-off Quote
Fiery Posted March 14, 2017 Posted March 14, 2017 Thanks a lot! Please try the latest AIDA64 beta available at: https://www.aida64.com/downloads/latesta64xebeta If it still crashes while scanning the SMBus, please let me know the new device index where it stops. Quote
jjxaker Posted March 15, 2017 Author Posted March 15, 2017 Problem still exists, already another address. And another problem: Quote
Fiery Posted March 15, 2017 Posted March 15, 2017 I've sent you an email about the SMBus issue. As for the benchmark reference results setting, that's exactly how it should work. Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.