Athlon X4 845 | Crossblade Ranger | And so it begins ...

[Massman]

Not much to say right now. I tried hard, but couldn't resist picking up an Athlon X4 845 Excavator chip to play SuperPI 32M with. Two weeks ago I went to the Guang Hua local computer market and saw it priced for ~ 2000 TWD. Next thing I remember was walking outside with a plastic back and a CPU.

 

I just installed the system and might do some testing this weekend. If you're upgrading from Trinity, Richland or Kaveri don't forget to FIRST flash the BIOS to the latest version! Excavator is a new architecture and it seems that without BIOS flash the board simply won't boot. Just a heads up ...

 

... now, has anyone tried a PCIe SSD to get around the BCLK frequency wall?

 

 

 

[flanker]

beat my score man:) (But seems, you must try Win7 for "auto" multiplier and CPB enabled.

[Lucky_n00b]

I smell 5G 32M AMD Platform will be crushed within weeks

[Massman]

beat my score man:) (But seems, you must try Win7 for "auto" multiplier and CPB enabled.

 

Yeah, that seems to be the main challenge right now. Can't get turbo ratio enabled in XP

[l0ud_sil3nc3]

Yeah, that seems to be the main challenge right now. Can't get turbo ratio enabled in XP

 

Please figure it PJ my board/cpu shows up Tuesday and will be ready to battle

[Massman]

From what I see, the issue is that in XP the CPU is stuck in P2 state and there is no ACPI support to enable turbo mode. I checked with AmdMsrTweaker, which has 15h family support, and confirmed the max available ratio is in fact 38x. However, Turbo mode is disabled. Elmor's am3_pcratio tool also indicates that the 38x ratio is available in XP ... just not configurable for P2 state.

 

So, all I need to figure out is how to get RWeverything working on my XP install and how to switch P-states.

 

FYI, here is AMD's tech documentation for those who are interested in going hardcore: http://support.amd.com/TechDocs/50742_15h_Models_60h-6Fh_BKDG.pdf

 

[speed.fastest]

Proposing to Asus to make XP ACPI, and the race start again

[Massman]

I don't think we should ask for a fix unless there's absolutely no way we can make it work ourselves haha

[l0ud_sil3nc3]

If XP doesn't work this garbage goes back to Newegg

[flanker]

or Win7

[l0ud_sil3nc3]

or Win7

Bench pi, on W7 . . . .

 

 

[flanker]

cause 35x XP is less than 38x in win7

[The Stilt]

Athlon X4 845 is a locked SKU so you cannot configure the maximum boosted multiplier for any other PState than Pb0. The programming conditions for each PState on Carrizo are =< the original FID and VID of the PState.

 

Carrizo is PITA to work with anyway. I can think of several ways to get around this using SMU, but the implementation would be pretty complex and I have not been able to test it since I don't have any of these chips available.

 

Also you cannot force the PState to switch to any boosted PState, since PState 0 command points to P0 instead of Pb0. Boosted PStates are not visible to the PStateCMD register, unless you set the "number of boosted states" value to zero (which cannot be done with ease).

 

It seems that BR will be useless too as it appears that AMD won't be releasing any unlocked SKUs

Edited May 17, 2016 by The Stilt

[Massman]

38x ratio is indeed working in XP, however it can't be forced and it seems to be not always kicking in at high voltages/bclk/oc (not sure what). In idle it will stick to 35x, so the boost kicks in after the process has started.

 

This isn't going to be an easy ride ...

 

@The Stilt, are you up for some testing with Carrizo? I'm totally up for shipping some parts if needed.

[The Stilt]

No need to send anything over as these chips can be had for 60€ or so. I just haven't had any interested on these since I already have a Carrizo in a laptop which I have already tested throughly.

 

So let say if you start SuperPI, the executing core (CU) will jump to 38x multiplier? If that's the case then Turbo is obviously active and working. Carrizo is the first AMD chip which can accurately monitor it's operating parameters and adjust the frequency accordingly. At some point you will most likely be limited by the TDP limit, at least when multiple cores are used. Even the mobile Carrizos running at significantly lower voltages require around 50W TDP to maintain all cores at 3400MHz during Cinebench.

 

I don't expect that the "number of boosted states" can be changed on this CPU. You would need to change it to zero in order to constantly use the highest available multiplier (38x). However if you're already limited by the TDP, then it won't obviously help much. The only other way around would be basically cheating the power management. If AMD hasn't disabled the TDP control (through SMU), increasing the TDP limit to "sufficient" levels would make the chip run constantly at the maximum frequency under the load. This works for mobile Carrizos at least.

 

Also the information displayed by MSRTweaker is wrong for most parts. The displayed voltages are wrong (SVI scale used instead of SVI2) and the other information is not displayed properly either. The multipliers are correct, but that's about it.

 

To know your original voltages:

 

Calculate the delta between the voltage displayed by MSRTweaker (each PState) and 1.55V. Divide the delta by two and add it to the displayed value. 1.40000V for "P0" (Pb0) is actually 1.47500V.

 

I'll let you know if I find a good way to solve the pending issue.

 

Edit: Check D18F4x15C.

 

Bit 31:31 is 1, correct (BoostLock)?

Bits 4:2 is 2, correct (NumBoostStates, Pb0 38x & Pb1 37x on this CPU)

 

If that's the case, then increasing the TDP is most likely the only way. Unless you can "tune" the engineer sandbox fuses...

Edited May 18, 2016 by The Stilt

[l0ud_sil3nc3]

No need to send anything over as these chips can be had for 60€ or so. I just haven't had any interested on these since I already have a Carrizo in a laptop which I have already tested throughly.

 

So let say if you start SuperPI, the executing core (CU) will jump to 38x multiplier? If that's the case then Turbo is obviously active and working. Carrizo is the first AMD chip which can accurately monitor it's operating parameters and adjust the frequency accordingly. At some point you will most likely be limited by the TDP limit, at least when multiple cores are used. Even the mobile Carrizos running at significantly lower voltages require around 50W TDP to maintain all cores at 3400MHz during Cinebench.

 

I don't expect that the "number of boosted states" can be changed on this CPU. You would need to change it to zero in order to constantly use the highest available multiplier (38x). However if you're already limited by the TDP, then it won't obviously help much. The only other way around would be basically cheating the power management. If AMD hasn't disabled the TDP control (through SMU), increasing the TDP limit to "sufficient" levels would make the chip run constantly at the maximum frequency under the load. This works for mobile Carrizos at least.

 

Also the information displayed by MSRTweaker is wrong for most parts. The displayed voltages are wrong (SVI scale used instead of SVI2) and the other information is not displayed properly either. The multipliers are correct, but that's about it.

 

To know your original voltages:

 

Calculate the delta between the voltage displayed by MSRTweaker (each PState) and 1.55V. Divide the delta by two and add it to the displayed value. 1.40000V for "P0" (Pb0) is actually 1.47500V.

 

I'll let you know if I find a good way to solve the pending issue.

 

Edit: Check D18F4x15C.

 

Bit 31:31 is 1, correct (BoostLock)?

Bits 4:2 is 2, correct (NumBoostStates, Pb0 38x & Pb1 37x on this CPU)

 

If that's the case, then increasing the TDP is most likely the only way. Unless you can "tune" the engineer sandbox fuses...

 

 

[Massman]

Edit: Check D18F4x15C.

 

Bit 31:31 is 1, correct (BoostLock)?

Bits 4:2 is 2, correct (NumBoostStates, Pb0 38x & Pb1 37x on this CPU)

 

If that's the case, then increasing the TDP is most likely the only way. Unless you can "tune" the engineer sandbox fuses...

 

D18F4x15C

[31:31] = 0

[4:2] = 2

 

D18F4x16C

[3:3] = 1 (TdpLimitDis)

 

[l0ud_sil3nc3]

D18F4x15C

[31:31] = 0

[4:2] = 2

 

D18F4x16C

[3:3] = 1 (TdpLimitDis)

 

 

So did it work? Solid x38 multi in XP while running pi?

[Massman]

Yea - just not force 38x in idle

[Massman]

Forget about what I just wrote. Disable two cores and 38x is active even in idle

[The Stilt]

D18F4x15C

[31:31] = 0

[4:2] = 2

 

D18F4x16C

[3:3] = 1 (TdpLimitDis)

 

 

There might be hope.

TdpLimitDis shouldn't do anything since the power management is a SMU feature. This bit should only affect Apm itself. Will writing 15Ch Bits 4:2 to 0 (from 2) stick?

[Massman]

There might be hope.

TdpLimitDis shouldn't do anything since the power management is a SMU feature. This bit should only affect Apm itself. Will writing 15Ch Bits 4:2 to 0 (from 2) stick?

 

Doesn't stick

 

[Massman]

First practice rounds ...

 

 

Edited May 18, 2016 by Massman

[Massman]

There's quite a bit of juice left in the trunk. Reminds me of Kabini - everything helps to boost PI

 

[filmbot]

So would the great fix be adjustable or removable TDP by the motherboard?

 

Am I gathering correctly that while we can force a PState in XP currently, we still have to be mindful of TDP as it will change the PState as we get closer or reach the TDP limit?