On Sun, 9/20/2026 5:08 AM, J. P. Gilliver wrote:
On 2026/9/20 1:43:28, Paul wrote:
On Sat, 9/19/2026 3:50 PM, J. P. Gilliver wrote:
On 2026/9/19 16:59:2, Paul wrote:
[]
SpeedFan is out of support, and when new SuperIO show up, the Hardware >>>> Monitor may not be supported by the hobby community. That's why the
Not sure how old this laptop is, but far from new. Let's see, Speccy
(yes, I know, but it does what I need) tells me it has a socket FP4
(968), Radeon R5, BIOS date 2017/4/17; My SpeedFan says it (SpeedFan) is >>> 2000-2016, so it might indeed not know the hardware.
There is a list of FP4 socket processors here, but I don't know
how complete this is. These are mostly relatively low power.
https://www.techpowerup.com/cpu-specs/?socket=AMD%20Socket%20FP4
You could try the ZIP-English here, as that might be the portable one.
If the screen gets covered up, do a Reload of the page. The first pane
of CPUZ shows processor details. The last button, has an option to
save a text file, if you want to pick bits out to make a text posting
with them. The text file has a lot of details that only a few people
are interested in (DIMM SPD contents and the like).
https://www.cpuid.com/softwares/cpu-z.html
Paul
Sorry - Speccy does tell me a lot about the CPU; I didn't mention it as
I was really only looking for the age of the laptop, to see if it was
later than SpeedFan to explain why the latter might not be able to see
its fans. Here's what Speccy says about the CPU:
AMD A10-9600P
Cores 4
Threads 4
Name AMD A10-9600P
Code Name Bristol Ridge
Package Socket FP4 (968)
Technology 28nm
Specification AMD A10-9600P RADEON R5, 10 COMPUTE CORES 4C+6G
Family F
Extended Family 15
Model 5
Extended Model 65
Stepping 1
Revision BR-A1
Instructions MMX (+), SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, SSE4A, AMD 64, NX, VMX, AES, AVX, AVX2, FMA3, FMA4
Virtualization Supported, Enabled
Hyperthreading Not supported
Bus Speed 99.8 MHz
Stock Core Speed 2400 MHz
Stock Bus Speed 100 MHz
Caches
L1 Data Cache Size 4 x 32 KBytes
L1 Instructions Cache Size 2 x 96 KBytes
L2 Unified Cache Size 2 x 1024 KBytes
Cores
Core Speed Multiplier Bus Speed Threads
Core 0 2594.7 MHz x 26.0 99.8 MHz APIC ID: 0
Core 1 2594.7 MHz x 26.0 99.8 MHz APIC ID: 1
Core 2 2594.7 MHz x 26.0 99.8 MHz APIC ID: 2
Core 3 2594.7 MHz x 26.0 99.8 MHz APIC ID: 3
(While I'm watching it, the core speeds shown are varying - mostly just
by a little, e. g. to 2395.1, occasionally to 1996.0, but only for a
moment [I had to wait some tens of seconds to catch that]. And only one
of the four changes - though as I typed that I saw two of them start
with 1 for one sample.)
OK, this is the information we need.
https://www.reddit.com/r/Amd/comments/5g270f/is_the_a109600p_26ghz_any_good/
"Also, I have the problem where my laptop gets too hot and restarts,
someone also mentioned this, so it might be minimal.(Blame the laptop fans
and HP's build)
"
That suggests the THERMTRIP is set too low. An Intel one might be set
at 110C or 120C or so, whereas the outboard (third party chip for safe shutdown)
those can be set at 85C. An AMD of that era, should have internal THERMTRIP just
like Intel, so it should not be artificially trimmed to 85C. My AthlonXP board had the outboard safety chip (added late in the design cycle, you had to check the revision on the motherboard before buying), and that chip it was hard to tell
exactly what it was really designed for. It was a small chip and not designed for infinitely variable trip point. (They could have used a couple of precision resistors to set the trip point.) The safety chip was included, to protect against
"my CPU cooler fell off" accidents :-) The processor would overheat instantly (2 seconds) and it could crack the die, the code running at the time would crash
and so on. Those were bare die days, and you could chip little pieces out
of the edge of the thing, without too much trouble. They used to include a shim to help protect the die, for some of those designs.
A Geekbench lists it as
A10-9600P
2400Mhz Base
3300Mhz (Turbo of sorts)
4C 4T 15W TDP <=== allowed to go higher for 80-100 seconds
Radeon R5 iGPU (which throttles when the CPU turbos)
and the multicore rating is only 2x the single thread, so
it's "pretty strangled" by circumstances beyond its control.
https://www.notebookcheck.net/Bristol-Ridge-in-Review-AMDs-A10-9600P-Against-the-Competition.168477.0.html
"While idling, the Pavilion 15-aw004ng runs at 5.7 to 9.3 watts"
And that's the level of heating it has to be able to withstand
in your hot weather. Yours is running at 26x multiplier while
the claimed lower limit is 24x. Yet some of the NotepadCheck benchmark
results find the CPU clock is lower than the 2400MHz number, it
does appear the multiplier can go lower. Throttling is an additional
effect possible in CPU design, where no-operation codes are injected
into the instruction stream to reduce power consumption.
Might need more fan... Or some hand-optimizing of the cooling path.
After all, every customer is a thermal engineer, right ? We live
for these opportunities to stamp out stupid stuff.
I've told you before, of my buying electronics having a "decorative" fan.
You could see the fan hole OK. But, they placed no intake holes on the other side of the housing for the thing. The result is the fan spins, but
it can't cool anything. My electric drill fixed the two items designed
that way. I must have put 20-30 holes in a front area, to provide
an intake vent. That is your job as the customer, to have your
electric drill handy, to finish the design. And I don't really enjoy
drilling hard plastics, as some of them prefer to crack rather than
be drilled.
Paul
--- PyGate Linux v1.5.19
* Origin: Dragon's Lair, PyGate NNTP<>Fido Gate (3:633/10)