• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD unleashes first ever commercial “5GHz” CPU, the FX-9590

Page 23 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
inf64 said:
We have one benchmark vs intel parts where it's among top performers (10% faster than 4770K @ stock; 4770K can OC ~10-15% so even OCed with increased power draw it can roughly only match new FX in this benchmark).
Click to expand...

You're looking at heavily threaded tests, where the 4770K (stock) will be running at base clock with no turbo. A typical Haswell OC will take you about 25-35% higher than its stock base clock, although it sounds like there are some duds that won't get this high..
 
Exophase said:
You're looking at heavily threaded tests, where the 4770K (stock) will be running at base clock with no turbo. A typical Haswell OC will take you about 25-35% higher than its stock base clock, although it sounds like there are some duds that won't get this high..
Click to expand...

Based on everything I've heard that is backwards - most hit 4.3, 4.5 at high volts, very few are able to go above 4.6.
 
sushiwarrior said:
Based on everything I've heard that is backwards - most hit 4.3, 4.5 at high volts, very few are able to go above 4.6.
Click to expand...

4.3 = 23% improvement
4.5 = 29% improvement
4.6 = 32% improvement

I don't see what's backwards about what I said (or why you'd use that word to be honest). Asus says 70% hit 4.5GHz for them, so I would qualify that as "typical."
 
joshhedge said:
That doesn't mean that a golden 83xx couldn't run at those frequencies at that voltage.
Click to expand...

No question there because it is pretty obvious where the 9590's are going to be coming from.

Just don't expect any of those golden samples to be left in the general population of retail 8350's anymore. They've all been mined out and binned as 9590's in advance of this announcement.
 
Idontcare said:
No question there because it is pretty obvious where the 9590's are going to be coming from.

Just don't expect any of those golden samples to be left in the general population of retail 8350's anymore. They've all been mined out and binned as 9590's in advance of this announcement.
Click to expand...

If AMD is really charging > $900 for 9590 they'll get sold as 8350s again sooner or later 😛
 
Part of me wanted to get a 8350 for a toy to replace the 1090 but now might as well wait. As IDC said if all the golden chips are already taken for higher parts might as well wait.
 
Exophase said:
If AMD is really charging > $900 for 9590 they'll get sold as 8350s again sooner or later 😛
Click to expand...

I have been thinking about the price. It seems to me AMD is the proverbial "between a rock and a hard place". If they are marketing it as this super uber performance chip, setting world records for stock frequency, etc., can they really sell it at a price in the range of Intel's high end mainstream chip? Would not this be admitting that it is really not an earth shattering performance breakthrough?

And if they do price it in the range of Intel's extreme edition cpus, they cannot match the performance, especially since the intel would have a lot of overclocking headroom left.
 
buying a higher clocked (700-800MHz) sub $200 chip for $900 makes a lot of sense 🙄

how can anyone choose this over the $600 3930K!?

but the price is probably wrong!?
 
frozentundra123456 said:
I have been thinking about the price. It seems to me AMD is the proverbial "between a rock and a hard place". If they are marketing it as this super uber performance chip, setting world records for stock frequency, etc., can they really sell it at a price in the range of Intel's high end mainstream chip? Would not this be admitting that it is really not an earth shattering performance breakthrough?

And if they do price it in the range of Intel's extreme edition cpus, they cannot match the performance, especially since the intel would have a lot of overclocking headroom left.
Click to expand...

IMO they should price it against SB-E similarly to how FX-8350 is priced against i7-4770K. Meaning more than i7-4770K and less than i7-3930K. Something like $400-450.

But I think they won't because this is too much of a limited edition product.

Does anyone else remember when JF-AMD used to talk about how halo products were pointless? Guess that is not what AMD actually thinks, at least not now...
 
MisterMac said:
Jesus christ - a stock 3930k is most "likely" faster.

Yet they price it ABOVE that?

I expect MOBO costs for it will also be more expensive due to VRM requirements than a cheap 2011 board.


Holy shit.


I geuss it's good there's people like galego.
He probably thinks it's only a 50% price increase for 100% performance.

😀


A 400 usd 9590 would have been KILLER enthusiast chip.
Click to expand...

It's only an O/C'd FX-8350, which is a $200 chip. Don't fall for the the hype.
 
3DVagabond said:
It's only an O/C'd FX-8350, which is a $200 chip. Don't fall for the the hype.
Click to expand...

Yeah I'm not getting how a factory overclocked CPU is somehow an "enthusiast" chip, have we really gotten that lazy?

The only interesting aspect of this release is the potential power decrease and possible peak clock potential increase from a tech standpoint.
 
galego said:
Testing performance is fine because an OC 8350 will work as a Centurion, but their measurement of power consumption is not, because the Centurion is not a OC 8350.
Click to expand...

It's going to use a lot of power. AMD isn't trying to hide it. They are listing the TDP as 220W We can settle any power debate real fast.

"Anyone who is concerned with perf/w need not apply."​
 
3DVagabond said:
It's only an O/C'd FX-8350, which is a $200 chip. Don't fall for the the hype.
Click to expand...


Then don't fall for the hype of the FX-8350 either. You might as well get an 8320. The point is these chips are guaranteed to hit 5ghz and some people will throw a little more coin for that guaranteed frequency. Some 8350s can't even sustain 4.7ghz and most 8320s can't.
 
Rvenger said:
Then don't fall for the hype of the FX-8350 either. You might as well get an 8320. The point is these chips are guaranteed to hit 5ghz and some people will throw a little more coin for that guaranteed frequency. Some 8350s can't even sustain 4.7ghz and most 8320s can't.
Click to expand...

They've been saving the good chips :thumbsup:
 
Dice144 said:
Part of me wanted to get a 8350 for a toy to replace the 1090 but now might as well wait. As IDC said if all the golden chips are already taken for higher parts might as well wait.
Click to expand...

In the past the chips they've done this with have been leaky chips that are worse in every respect except if you can give them more volts and increase clocks. Remember the TWKR chips?

If you really want an FX-8350 to play around with, go buy it with the confidence that these leaky chips have been weeded out. Then sit back, eat some popped corn, and enjoy the antics of people who paid 3x to 5x as much you did for lower binned chips.
 
AMD has to guarantee that 5ghz operation. I believe that's why the TDP is so high. They have to go with probably the best choice cpu's and high voltages. The simulation reviews didn't even get 5ghz and we don't know if they are 24/7 stable, which is what the new cpu's will be. The real reviews will be a good read.
 
Tsavo said:
Almost $1000 for this thing.

AMD needs a company wide anti drugs policy.
Click to expand...

Haha that is the first thing that came to mind.

I think i would rather pay $900 for a

Intel® Core™2 Extreme Processor QX9775

it would give about the same performance 🙂
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top