CPUZ 1.73 Benchmark Thread

[LTC8K6]

Maybe I'm overlooking something....Why is that scoring so much lower than soccerballtux?

http://www.cpu-world.com/Compare/305/AMD_FX-Series_FX-8150_vs_AMD_FX-Series_FX-8310.html

 

[Edgemeal]

FX8150 @ 4.6GHz
win 10 64bit

Why such a huge score difference between these two?

[B]
  Name           | Score  | CPU      | MHz[/B]
-----------------+--------+----------+-----
1 [URL="http://forums.anandtech.com/showpost.php?p=37629747&postcount=4"]soccerballtux[/URL]  | [URL="http://i.imgur.com/txH92i2.png"] 6642[/URL]  | FX-8310  | 4602
2 [URL="http://forums.anandtech.com/showpost.php?p=37640427&postcount=153"]AtenRa[/URL]         | [URL="http://i60.tinypic.com/11h9npv.jpg"] 4461[/URL]  | FX-8150  | 4613

 

[AtenRa]

Bulldozer vs Vishera maybe ??? I will try my 8350 when ill finish benchmarking my 8150.

 

[SlowSpyder]

Maybe I'm overlooking something....Why is that scoring so much lower than soccerballtux?

For some reason, this bench seems to be extra hard on Bulldozer vs. Piledriver. Vishera seems to do better than the newer APU's too (L3?).

 

[Dresdenboy]

Bulldozer vs Vishera maybe ??? I will try my 8350 when ill finish benchmarking my 8150.

This is likely the same reason as for this jump in the Passmark integer results:

from http://www.planet3dnow.de/cms/18564...cavator-leistungsvergleich-der-architekturen/

It's simply the availability of an integer HW divider!

 

[YBS1]

This is likely the same reason as for this jump in the Passmark integer results:

from http://www.planet3dnow.de/cms/18564...cavator-leistungsvergleich-der-architekturen/

It's simply the availability of an integer HW divider!

For great science!

 

[AtenRa]

This is likely the same reason as for this jump in the Passmark integer results:

from http://www.planet3dnow.de/cms/18564...cavator-leistungsvergleich-der-architekturen/

It's simply the availability of an integer HW divider!

Hell, i havent seen this one. thanks

 

[SPBHM]

It's simply the availability of an integer HW divider!

so... anything older than Piledriver for AMD, and older than C2D (?) for Intel is in trouble for this?

 

[Dresdenboy]

so... anything older than Piledriver for AMD, and older than C2D (?) for Intel is in trouble for this?

It's one big differentiator between BD and PD or K10 and K10.5. There are other factors of course, which gain weight as the idiv's one diminishes.

 

[LTC8K6]

i3-4360 1350/2350

Today it's 1410/2372.

Only difference is going from 8gb of ram to 12gb of ram.

 

[LTC8K6]

FX-6300 1866 CL9 ram

875/3741

 

[Edrick]

6600K @ 4.2ghz

 

[Burpo]

Thank you Edrick! Single thread beats 4790K ..

 

[LTC8K6]

Thank you Edrick! Single thread beats 4790K ..

Meh, wonky bench and not enough of a difference to matter.

1715 / 6261

4790K 4.4 2133 ram

 

[SlowSpyder]

I just got Windows 10 and don't have the media, so this is about as far as I care to go... don't want to bork anything and have to go back to Win7. I'm sure there are faster FX's, but I have to be in the 99th percentile. :awe:

5.278GHz (203 x 26) - 1251 / 7837

 

[DrMrLordX]

Two updates:

1). Unlike an earlier poster that claimed that 1.73 wouldn't work in Win10 build 10525, I have been able to get it to run. I got a higher score, too. My latest score is 805/2366, which is slightly higher than my previous score of 785/2356

2). I tried it on an E1-2500 @ 1.4 GHz. Its score was 341/282. Ouch!

 

[LTC8K6]

Two updates:

1). Unlike an earlier poster that claimed that 1.73 wouldn't work in Win10 build 10525, I have been able to get it to run. I got a higher score, too. My latest score is 805/2366, which is slightly higher than my previous score of 785/2356

2). I tried it on an E1-2500 @ 1.4 GHz. Its score was 341/282. Ouch!

I said that Windows said it was not compatible, not that it didn't work.

When 10525 installed, it gave the notification about CPU-Z, and uninstalled it.

I had to reinstall it.

W10 10525 also uninstalled a network driver it didn't like.

All of the benches I posted were running W10.

But the bench is beta, and needs to be fixed.

 

[Dufus]

Just got a G3258 to experiment on but unfortunately have to go overseas for a while.

Just a quick run at ~1.47V, 2 cores at 29x

Noticed something odd with MT scores when affinity is set to threads -1 with 2C/4T CPU. Maybe someone can confirm.

 

[richierich1212]

Nice Chip Dufus!! Mine won't boot into Windows at 50x. 49x takes 1.55+vcore and still isn't stable enough for the bench:

 

[jihe]

X5650@ 189*22 = 4.15G

Seems to scale linearly with physical cores

 

[soccerballtux]

I just got Windows 10 and don't have the media, so this is about as far as I care to go... don't want to bork anything and have to go back to Win7. I'm sure there are faster FX's, but I have to be in the 99th percentile. :awe:

5.278GHz (203 x 26) - 1251 / 7837

yowser, very nice

 

[know of fence]

Just got a G3258 to experiment on but unfortunately have to go overseas for a while.

Just a quick run at ~1.47V, 2 cores at 29x

Noticed something odd with MT scores when affinity is set to threads -1 with 2C/4T CPU. Maybe someone can confirm.

Very nice, can someone explain what is going on with Core Voltage.

 

[Dufus]

C-States probably.

The voltage is read from the Super IO rather than reading the VID (which is ~30mV lower BTW). When the CPU is idle clocks are stopped and voltage gating may take place, check C6 power state in Intel datasheets. Monitoring software wakes the CPU and reads clocks but there may be insufficient time for the SIO to poll the ADC so a stale reading is taken. Either that or something really wrong elsewhere .

Anyway, here's a snapshot during the bench. Note it can complete at lower voltage but I prefer some safety margin for stability. Snapshot taken during MT at 29x, ST was at 30x. Using fixed voltage FWIW, bclk strap messes up voltage / VID scaling.

 

[DrMrLordX]

athlon2 x4 620
single: 382
multi: 1039

Yeouch. I know the clockspeeds aren't that high and there's no L3 but still. The ST score isn't that far ahead of the E1-2500 . . .

I said that Windows said it was not compatible, not that it didn't work.

When 10525 installed, it gave the notification about CPU-Z, and uninstalled it.

I had to reinstall it.

Oh okay. That didn't happen to me either. Of course I'm not using the "installed" version, just the binary from the .zip distribution. I never use the installer.

 

[LTC8K6]

Yeouch. I know the clockspeeds aren't that high and there's no L3 but still. The ST score isn't that far ahead of the E1-2500 . . .



Oh okay. That didn't happen to me either. Of course I'm not using the "installed" version, just the binary from the .zip distribution. I never use the installer.

It's an insider test mule. I built it back in September to use on the W10 insider program. It has had all of the builds auto-installed since then. They may not have wanted the software on there with the release builds, since I was a beta tester?