• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

AMD FX-9590 - in FlanK3rs hands and in review

FlanK3r

FlanK3r

Joined
May 18, 2009
Messages
406 (0.07/day)
Location
czech republic
System Specs
System Name AMD forever, AMD overclocker
Processor Athlon 3000+ Venice, Athlon x2 4600+ EE Windsor,x4 955 BE@3.9GHz AIR, X4 965 BE, x4 970 BE, x6 1090T
Motherboard Asus C5F-Z, Asus C5F, Asus C4E, Asus C4F, Asus RIIIE, Asus R4E, Asus M5F, Asus M6F, Asus M7H,
Cooling Corsair H100, Swiftech H220, CM Hyper 212, Xigmatek 1283 DK+Ultra Kaze, CM V6GT, Noctua NHD14
Memory Kingston Hyper X, 1600 A-Data 2000x, Corsair Dominator GT 2000 MHz, GSkill TridentX 2400MHz, HyperX
Video Card(s) HD3870 512MB GDDR4, HD5770 1024MB GDDR5, ATI HD4870, HD 6870 GDDR5, HD 7870,Radeon R9-270X
Storage 2x 320GB WD+Samsung, 1x 500 GB Samsung, SSD X-25, SSD HyperX, SSD Seagate, SSD Corsair GT
Display(s) BenQ 24" 24XL
Case HAF 922, Aspire X-Cruiser, benchtable Wroom
Power Supply Seasonic 500W and Seasonic 650W, Corsair AX 1200W, Zalman Goldrock 750, Galaxy OC edition 1200W
Software x64 win 7 and x86 Win XP SP3
Benchmark Scores you rather I do not want to see .. :-D !!!Thuban coming soon, new secret Gigabyte mobo too !!!
AMD FX-9590 ES review at PCTUNING (use translate from czech language to english). At the beginning I wanna tell two things.
1) Intel CPUs has 1600 MHz DRAm as default IMC settigs, AMD 1866-2133, depends at model.
2)Intel Z87/77 motherboards have set sometimes wrong Intel stock Turbo, so I set it right and manually at 39 (one)-38 (two)-38 (three)-37(for loading) for cores (3770k+4770k)

http://pctuning.tyden.cz/hardware/p...rocesor-se-stava-realitou-amd-fx-9590?start=1

ScreenShot103.png


ScreenShot106.png


ScreenShot116.png


ScreenShot117.png


ScreenShot101.png


ScreenShot121.png


ScreenShot120.png


ScreenShot137.png


ScreenShot098.png


ScreenShot099.png


ScreenShot114.png


ScreenShot113.png


ScreenShot115.png


ScreenShot131.png


ScreenShot112.png


ScreenShot122.png


For graphs: "méně je lépe" = "lower value is better" "více je lépe" = "higher value is better"
 
Is there not something intrinsically wrong with that set up?

The Haswell is teamed with a 7870 for all tests. But we get to see Vishera with a 7990. I appreciate it might be for demonstration of ability but it's not helpful to compare two cpu's when one is running AMD's 4th/5th most powerful gpu and the other is running with AMD's most powerful gpu.

Unless i'm missing the point?
 
I too would call foul, but all the tests shown are CPU only, not GPU...
 
most of the other reviews ive seen have intel in favour.

i hope discussion here will shed some light on this, or other reviews.

different RAM speeds and timings on different platforms??
 
Last edited:
The problem is, it might beat the intels in some of these tests AT STOCK, but this isnt going to overclock much more, whereas the intel has a buttload of overclocking headroom to go. Not to mention it consumes twice as much power.
8350 was an awesome CPU, this thing is just a polished turd.
 
RCoon said:
The problem is, it might beat the intels in some of these tests AT STOCK, but this isnt going to overclock much more, whereas the intel has a buttload of overclocking headroom to go. Not to mention it consumes twice as much power.
8350 was an awesome CPU, this thing is just a polished turd.
Click to expand...

reading that article and it says

''Disappointed by the inability to reach a 5 GHz on all cores I tried at least tests Superpi.''

(and somethign about the motherboard may be to blame and temps)

nothing more than cherry picked and overclocked 8350s

ive seen 8350s do a lot better....
 
d1nky said:
most of the other reviews ive seen have intel in favour.

i hope discussion here will shed some light on this, or other reviews.

different RAM speeds and timings on different platforms??
Click to expand...

yes, because we talking about stock IMS settings. Haswell+IB has stock 1600 MHz IMC, Trinity, Vishera has 1866 MHz as CPU default IMC. Richland has 2133 MHz for A10. Sandy Bridge, Phenoms II exmaple have 1333 MHz...

Of course, this CPU can run DRAM higher, but we "must" holt Intel and AMD specification. The same for Intel turbo. Motherboard vendors dont help with it. They have "edhance turbo spec", no Intel specification for turbo mode. This is 39-38-38-37 turbo multiplier for i7-4770K. But many motherboards was tested in reviews wrong with 39-39-39-39. Its "cheat" from my point of view.

HD7990 vs HD7870.
The PC demonstration was ES FX-9950 in full build PC system. I cant changed it. I wrote in text, we can not compare GPU results and power consumption of others PC. But I focused at CPU benchmarks and real aplications.
 
It looks to me that it is performing comparably, in benchmarks and that is better than previous fx series chips which is good news
 
i see, well in that case most reviews are biased by using non stock settings. thanks.

and what a shame about teh overclocking, will you be putting it on LN2/dice?
 
the second one, retail FX 9950, yes. But in August or so...
 
erocker said:
Not bad!

I don't understand why for the benchmarks, they clock the intel chip to 4.4ghz but for the power consumption test, they have it at 4.8ghz. That's a strange way to test...
Click to expand...
power-consumption2.png

cinebench115.png


You're looking at the 3960X? It seems that their comparisons were from tests a while ago and clock speeds vary. There are 3960X comparisons at 4.8, perhaps not for all though. Nice catch.
 
erocker said:
Not bad!

I don't understand why for the benchmarks, they clock the intel chip to 4.4ghz but for the power consumption test, they have it at 4.8ghz. That's a strange way to test...
Click to expand...

Yeah, comparing 4.4 GHz 3960X and STOCK 9590 is...dumb. OC the 9590, too, thanks, bbqstew.:cry:
 
Looking around at reviews this chip is not a golden clocker which means it's got nothing going for it over a 8350 or any other proc. It's only hope of holding off Intel and justifying the price was if it could hit like 5.8+ ghz on top end cooling. So I'm left wondering what the point is. It's not like they took any crowns back or tempted any buyers away from Intel ED chips.
 
we'll have to see what it does on LN2 or dice i guess, maybe itll break a new record.

if it doesnt, then i dont see the point in it.
 
Why is the 4770K scoring 1,000 points lower in 3DMark11 than the 3770K?
 
Is this review done at its stock clocks of 4.7GHz all 8 cores with turbo set at 5GHz four cores?
 
yes, it is. 4.7 GHz and up to 5 GHz turbocore 3.0

Xaser04: I dont know, maybe is the reason this. 90% of reviews in D-day reviewed Haswell and motherboards with non Intel turbo...With edhanced turbo. This mean, all stress time (no depends if 1 core or 8 logical cores) run at 3900 MHz. So +200 Mhz more than Intel Turbo specification. This gain can help in many hard load benchmarks. The second reason could be DRAM clock. Haswell scaling good with example 2400 MHz DDR3. But again, the default IMC in Intel spec is 1600 MHz without XMP.
 
FlanK3r said:
Xaser04: I dont know, maybe is the reason this. 90% of reviews in D-day reviewed Haswell and motherboards with non Intel turbo...With edhanced turbo. This mean, all stress time (no depends if 1 core or 8 logical cores) run at 3900 MHz. So +200 Mhz more than Intel Turbo specification. This gain can help in many hard load benchmarks. The second reason could be DRAM clock. Haswell scaling good with example 2400 MHz DDR3. But again, the default IMC in Intel spec is 1600 MHz without XMP.
Click to expand...

I might be miss-reading what you are saying here, but wouldn't this lead to a more favourable score for the 4770K?
 
cdawall said:
It depends on the test for clock speeds. I really wish someone would really push the 9590 already :laugh:
Click to expand...

This was from kitguru's review.
"The system was prime stable at these settings and the core voltage showed 1.476V in the BIOS and around 1.49V in CPUz. The validation below shows 1.58 volts as we were pushing it harder to try further overclocking above 5ghz (unsuccessfully), but it was stable at 5ghz with 1.488v as shown in the CPUz screenshot below."

They tried to push it but the cooling solution was inadequate. These chips clock higher with less volts the cooler you can manage to keep them The numbers I'm seeing look just like the FX8350 so in theory they've already been pushed. I highly doubt that any of these will go any higher than an 8350 would.
Here's a good thread from OCF shows the OC'ing of an 8350 from 4.1 to 5.4. it was done for performance comparisons at clock speeds. Notice that at 5.0 the 835 was at 1.5v very similar ti the 9590. http://www.overclockers.com/forums/showthread.php?t=724953
 
8350 is a higher binned 8320... 9370...higher binned 8350...9590 higher binned 9370...

and only some high end boards handle the TDP of a 5ghz 8350...
 
Can those CPUs take 1.6v-1.7v on water?

For more than a benchmarking session, I mean.
 
Mine takes 1.6v all day long, when my ambients arent 35*c LOL
 
You must log in or register to reply here.
Back
Top