AMD Firmware Update To Bring Boost Performance Optimizations

[barti_ddu]

The whole Ryzen 3000 launch has been a myriad of screw-ups so far.

I am very content with my 3600 (non-X) so far. It boosts up to 4.1 GHz (all cores) and I've never seen nor 1.5V nor 20W power consumption at idle. I can provide more precise figures when I get home, however at least turbostat does report ~6-8 W at idle (AFAIR). T_die temperature doesn't seem to go over ~65°C so I'm not worried about chip longevity.

In general it is ~20-30% faster than 2600 which I had previously and handles memory much better.

However, I'm on old x370 chipset, that might explain why I don't have any problems You've mentioned.

[barti_ddu]

Have You tried plugging the keyboard in another USB port (preferably USB2 if exists)? Asrock is weird in this regard.

[birdie]

I am very content with my 3600 (non-X) so far. It boosts up to 4.1 GHz (all cores) and I've never seen nor 1.5V nor 20W power consumption at idle. I can provide more precise figures when I get home, however at least turbostat does report ~6-8 W at idle (AFAIR). T_die temperature doesn't seem to go over ~65°C so I'm not worried about chip longevity.

In general it is ~20-30% faster than 2600 which I had previously and handles memory much better.

However, I'm on old x370 chipset, that might explain why I don't have any problems You've mentioned.

And how do you check that under Linux exactly? Only HWINFO64 and Windows utilities alike show this info. I would like you to post a HWiNFO64 screenshot with CPU sensors on it, so that I won't have to call you a liar.

[Pseus]

The whole Ryzen 3000 launch has been a myriad of screw-ups so far.

• CPUs not hitting their advertising boost clocks (not even a single core)
• Buggy X570 BIOSes
• X570 motherboards "feature" horrible chipset coolers and insane chipset power consumption (still no comment from AMD on this one)
• Weird/very high CPU idle power consumption (~20W at idle as reported by HWInfo64)
• Weird/very high idle/load CPU voltages - 7nm transistors @ 1.5V voltage ... what could go wrong? Oh, Ryzen 3000 CPUs are already dying.

Oh, and we've had Linux specific issues, like

• No ACPI tables for Linux (check for ACPI: [Firmware Bug]: BIOS _OSI(Linux) query ignored in dmesg)
• the k10temp driver not working at all (probably fixed in yet to be released kernel 5.3)
• motherboard sensors not working due to ACPI region conflicts (to which lm-sensors developers say you must use a kernel boot option acpi_enforce_resources=lax - wow! as if average people even know how to do that)
• lower multithreaded performance than in Windows 1903
• a lot higher power consumption than in Windows (no idea when and if that'll be fixed)

However Linux users might not care because GeekBench shows that barely anyone uses Linux to run Ryzen 3000 CPUs - from what I've seen less than 0.5% of people submit GeekBench results under Linux. It speaks volumes about Linux adoption but I'm pretty sure quite a few apologists will reply to this message and explain to me how and why I'm wrong and how Linux has over 10% on desktop.

What a fine day!

I have an X570 board and a 3600X from day one:

- CPUs not hitting their advertising boost clocks (not even a single core)
This is true. My 3600X hits 4300 with no problems, but fails to hit 4400. Big deal, once they fix this I'll get my 2,3% missing performance.

- Buggy X570 BIOSes
The biggest bug I experienced was slow graphics in the BIOS with CSM disabled. This has been fixed since then. I have no other issues.

- X570 motherboards "feature" horrible chipset coolers and insane chipset power consumption (still no comment from AMD on this one)
I'm sitting at the PC at this moment. The chipset cooler is off. It only spins during POST. I have an NVMe drive, and 4 SATA disks. I guess it would spin if I had two NVMe PCI-E 4 drives. When it does spin, I can hardly hear it.

- Weird/very high CPU idle power consumption (~20W at idle as reported by HWInfo64)
Not sure about this, I use Linux, no HWInfo64. I can say, though, that my power bill has been unchanged since the last month.

- Weird/very high idle/load CPU voltages - 7nm transistors @ 1.5V voltage ... what could go wrong? Oh, Ryzen 3000 CPUs are already dying.
AMD confirmed that 1.5v are normal under low current scenarios. In high current scenarios voltage is lower, and that works fine. I have seen no reports of unusually high number of failures.

- No ACPI tables for Linux (check for ACPI: [Firmware Bug]: BIOS _OSI(Linux) query ignored in dmesg)
Err, this has been the case for ages in both Intel and AMD platforms. My 4790K mobo shows the same.

- the k10temp driver not working at all (probably fixed in yet to be released kernel 5.3)
Yes, this is a new CPU, you need to wait until kernel developers add support for its features, like temperature monitoring. This has always been the case.

- motherboard sensors not working due to ACPI region conflicts (to which lm-sensors developers say you must use a kernel boot option acpi_enforce_resources=lax - wow! as if average people even know how to do that)
That is a motherboard and motherboard BIOS issue. BTW, several motherboard manufacturers use Super IO chips (used for hardware monitoring) that are simply not supported under Linux, regardless of that option (which is a bigger issue). Guess what? Those chips are also used on Intel motherboards (like Gigabyte's Z390 series). Hardly an AMD specific issue, or even a Ryzen 3000 issue.

- a lot higher power consumption than in Windows (no idea when and if that'll be fixed)
Yes, this is a new CPU, you need to wait until kernel developers add support for its features, like new hardware sleep states. This has always been the case.

I've been happily using this machine for professional work since release. No issues whatsoever. The biggest one was the RDRAND bug, which you failed to mention and was an actual bug. Fortunately, it was a very easy workaround. Other than that I've had no issues. Pretty impressive, IMO, for a new platform and CPU.

If you are not satisfied, you can always buy something else.

[thelongdivider]

I have an X570 board and a 3600X from day one:

- CPUs not hitting their advertising boost clocks (not even a single core)
This is true. My 3600X hits 4300 with no problems, but fails to hit 4400. Big deal, once they fix this I'll get my 2,3% missing performance.

- Buggy X570 BIOSes
The biggest bug I experienced was slow graphics in the BIOS with CSM disabled. This has been fixed since then. I have no other issues.

- X570 motherboards "feature" horrible chipset coolers and insane chipset power consumption (still no comment from AMD on this one)
I'm sitting at the PC at this moment. The chipset cooler is off. It only spins during POST. I have an NVMe drive, and 4 SATA disks. I guess it would spin if I had two NVMe PCI-E 4 drives. When it does spin, I can hardly hear it.

- Weird/very high CPU idle power consumption (~20W at idle as reported by HWInfo64)
Not sure about this, I use Linux, no HWInfo64. I can say, though, that my power bill has been unchanged since the last month.

- Weird/very high idle/load CPU voltages - 7nm transistors @ 1.5V voltage ... what could go wrong? Oh, Ryzen 3000 CPUs are already dying.
AMD confirmed that 1.5v are normal under low current scenarios. In high current scenarios voltage is lower, and that works fine. I have seen no reports of unusually high number of failures.

- No ACPI tables for Linux (check for ACPI: [Firmware Bug]: BIOS _OSI(Linux) query ignored in dmesg)
Err, this has been the case for ages in both Intel and AMD platforms. My 4790K mobo shows the same.

- the k10temp driver not working at all (probably fixed in yet to be released kernel 5.3)
Yes, this is a new CPU, you need to wait until kernel developers add support for its features, like temperature monitoring. This has always been the case.

- motherboard sensors not working due to ACPI region conflicts (to which lm-sensors developers say you must use a kernel boot option acpi_enforce_resources=lax - wow! as if average people even know how to do that)
That is a motherboard and motherboard BIOS issue. BTW, several motherboard manufacturers use Super IO chips (used for hardware monitoring) that are simply not supported under Linux, regardless of that option (which is a bigger issue). Guess what? Those chips are also used on Intel motherboards (like Gigabyte's Z390 series). Hardly an AMD specific issue, or even a Ryzen 3000 issue.

- a lot higher power consumption than in Windows (no idea when and if that'll be fixed)
Yes, this is a new CPU, you need to wait until kernel developers add support for its features, like new hardware sleep states. This has always been the case.

I've been happily using this machine for professional work since release. No issues whatsoever. The biggest one was the RDRAND bug, which you failed to mention and was an actual bug. Fortunately, it was a very easy workaround. Other than that I've had no issues. Pretty impressive, IMO, for a new platform and CPU.

If you are not satisfied, you can always buy something else.

This dude is a troll. The x570 launch has been very welcomed by most in the industry. Every new chipset and CPU release has a few teething issues, but they will eventually get resolved. 12 core equal to greater IPC chips being sold at the same price as the 9900k? It's amazing. I can't wait to get my hands on the 3950x sometime this month.

[numacross]

This dude is a troll. The x570 launch has been very welcomed by most in the industry. Every new chipset and CPU release has a few teething issues, but they will eventually get resolved. 12 core equal to greater IPC chips being sold at the same price as the 9900k? It's amazing. I can't wait to get my hands on the 3950x sometime this month.

9900K doesn't come with a cooler and requires a beefy one while Ryzen 3900X comes with adequate (albeit not super-silent) cooler in box. It also works even on cheap A320 motherboards with RAM overclocking (which require a Z-series on Intel). The entire platform cost looks to be lower for AMD while delivering adequate (ST) or superb (MT) performance.

[thelongdivider]

9900K doesn't come with a cooler and requires a beefy one while Ryzen 3900X comes with adequate (albeit not super-silent) cooler in box. It also works even on cheap A320 motherboards with RAM overclocking (which require a Z-series on Intel). The entire platform cost looks to be lower for AMD while delivering adequate (ST) or superb (MT) performance.

I agree except for the fact that it has no a320 support. But why would anyone buy a motherboard like that anyways?

[numacross]

I agree except for the fact that it has no a320 support. But why would anyone buy a motherboard like that anyways?

Might not have official support, but it works somewhat My point was the AMD AM4 platform is less limited/segmented than Intel's equivalents.

[sa666666]

It speaks volumes about Linux adoption but I'm pretty sure quite a few apologists will reply to this message and explain to me how and why I'm wrong and how Linux has over 10% on desktop.

I would like you to post a HWiNFO64 screenshot with CPU sensors on it, so that I won't have to call you a liar.

You really are a condescending, passive-aggressive prick, aren't you? And a fairly poor troll, to boot. You have a chip on your shoulder that I can see from another country.

[Templar82]

Oh, and we've had Linux specific issues, like

You forgot the best one,

You couldn't even boot if you were using a Systemd distro