• 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.

Found another one: Ukraine store accidentally ships FX-8120 and it gets tested!

Page 7 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
mosox said:
That's 93% not 48%.
Click to expand...

2600k performs the task 48% quicker than FX-8150, or FX-8150 is 93% slower in this task than the 2600k.

From a perspective of the 2600k:

(New - Old) / Old * 100% - 1 = (10.64 seconds - 20.545) / 20.545 * 100 = - 48% or 48% faster than FX-8150

From a perspective of the FX-8150:

(20.545 - 10.64) / 10.64 * 100% = 93% slower than 2600k.
 
NostaSeronx said:
Gross, I would never want $1000 CPUs
Click to expand...

Why is it that AMD has to bench against a 1000 dollar intel cpu?

Seriously... Why are u guys even ranking it with them?

Why?

OH its because its the ONLY thing it will win...


so yeah, my Parents AMG S65 is faster then Acura TL back when i was in college..

But the TL was 1/4th the price of the S65!!! so the TL is obviously a better car... because it was cheaper.. even tho the S65 has more HP, a Larger Engine.. More Room... and is 2x more loaded with luxury features.

So you really think comparing a TL against a S65 is smart?
This is the type of analogy u give us when you compare a 200 dollar AMD processor against a 1000 dollar intel processor.

YOU JUST LOOK STUPID TO ANYONE WHO KNOWS CPU's... AND NO YOUR NOT CONVINCING ANYONE AMD IS A BETTER PROCESSOR BY RANKING THEM WITH INTEL'S 1000 DOLLAR PROCESSOR.


This is what u guys are doing..

What next? why not just throw in the most expensive hexcore xeon on Intel side instead..

The X5680... its only $1499.

Compare that instead of the 990X...
AMD = WiN because u can get 7 Bulldozers for 1 X5680.. and 7 x 8 = 56 cores while that single intel is 12 cores... :O


Please....

<sigh>
 
Last edited:
AMD can't beat Intel in milking more from the current architecture, they HAD to come up with something new. If you can't beat them in steam locomotives, try diesel locomotives.

The thing is all this is new and it's hard to make it great the first time. But if they did (completely new architecture AND performance crown) that would have meant they have the most awesome engineers ever.

FX is a good name for what they've got. Experimental CPUs. We try, you buy. Or not.
 
Medu said:
They already have a 32nm Thuban core- it's called Llano and we all know how good that CPU part works...
Click to expand...

mosox said:
AMD can't beat Intel in milking more from the current architecture, they HAD to come up with something new. If you can't beat them in steam locomotives, try diesel locomotives.
Click to expand...

Llano is not a full-fledged Phenom II core. It's just a slightly revised Athlon II X4 CPU. So it wouldn't perform as well as a 32nm Phenom II core. Also, a portion of Llano's die is dedicated to a GPU. Theoretically speaking, AMD engineers could have added more cache to Phenom II in that "GPU dedicated space space" or used that additional transistor space to improve the complexity of each original Phenom II core. Hindsight 20/20, but if they focused on improving IPC for Phenom II by 15-20&#37;, kept Phenom II at 4 and 6 cores and not pushed for 8 cores, and used 32nm to increase clocks to 4.2ghz, this chip would have been flying. Even with 0 changes to Phenom II and just increased clock speeds to say 4.0-4.2ghz (with 4.5-4.8ghz overclocking headroom), and the same 6% increase in IPC that Llano has over Athlon II X4, that BD would have been an awesome CPU.

Some user posted:

"My X6 1055T @3.6GHz at SuperPi 1M can hit 16.240s, but at that review, SuperPi 1M @5GHz only 17.019s
At Cinebench R11.5 64-bit, my X6 1055T @3.6GHz can hit 6.45 Pts, but at that review FX-8150 @default (3.6GHz) only 6.01 pts
And at wPrime v1.55, my X6 1055T @3.6GHz can hit 7.191s, and at that review only 8.845s with FX-8150 @default clock"

mosox said:
FX is a good name for what they've got. Experimental CPUs.
Click to expand...

What do you mean? FX was better in:
- power consumption
- overclocking
- performance per clock and per core
- absolute overall performance
- Not a single Intel CPU could touch the FX

How is this CPU worthy of the FX brand?

Maybe BD will perform much better under Windows 8 if it shares resources more efficiently than Windows 7 does, per AMD slides.
 
Last edited:
RussianSensation said:
Some user posted:

"My X6 1055T @3.6GHz at SuperPi 1M can hit 16.240s, but at that review, SuperPi 1M @5GHz only 17.019s
At Cinebench R11.5 64-bit, my X6 1055T @3.6GHz can hit 6.45 Pts, but at that review FX-8150 @default (3.6GHz) only 6.01 pts
And at wPrime v1.55, my X6 1055T @3.6GHz can hit 7.191s, and at that review only 8.845s with FX-8150 @default clock"
Click to expand...

and how long have i been tooting the X6 on the forum during BD launch?

And how many people shot me down saying i was trolling?

Where are the trolls now?


Sorry every way i look at it.. the X6 is a more superior CPU... Hands down... if its performance ur after for a budget.
 
RussianSensation said:
2600k performs the task 48% quicker than FX-8150, or FX-8150 is 93% slower in this task than the 2600k.

From a perspective of the 2600k:

(New - Old) / Old * 100% - 1 = (10.64 seconds - 20.545) / 20.545 * 100 = - 48% or 48% faster than FX-8150

From a perspective of the FX-8150:

(20.545 - 10.64) / 10.64 * 100% = 93% slower than 2600k.
Click to expand...

You are using time and speed interchangeably, this is wrong. Saying something is 93% slower than something else implies that it is moving at 7% of the speed of the faster thing. That is clearly not true. It is correct to say that FX-8130 is taking 93% more time than 2600K, or that 2600K only takes 48% of the time FX-8130 does.

Speed is the inverse of time, so to get speed numbers you need to first take inverses of the time values:

2600K "speed" = 1/10.64 = 0.094
FX-8130 "speed" 1/20.545 = 0.049
So FX-8130 is 48% slower than 2600K, or 2600K is 93% faster than 2600K.
 
RussianSensation said:
2600k performs the task 48&#37; quicker than FX-8150, or FX-8150 is 93% slower in this task than the 2600k.

From a perspective of the 2600k:

(New - Old) / Old * 100% - 1 = (10.64 seconds - 20.545) / 20.545 * 100 = - 48% or 48% faster than FX-8150

From a perspective of the FX-8150:

(20.545 - 10.64) / 10.64 * 100% = 93% slower than 2600k.
Click to expand...

Not really.

Old car 100 mph

New car 200 mph

New car twice as fast (100% faster), not 50% faster (as in 100 = 50% of 200 mph).
 
Last edited:
Tuna-Fish said:
You are using time and speed interchangeably, this is wrong.
Click to expand...

If Person 1 runs 100 meter dash in 10 seconds and Person 2 run it in 20 seconds, then:

1) Person 1 runs 100 meter dash in half the time, or completes the task 50&#37; faster than Person 2 (i.e., (10-20)/20*100 = 50% faster).

2) Person 2 runs 100 meter dash in double the time, or completes the same task 100% slower than Person 1 (i.e., (20-10)/10*100% = 100% slower).

No?
 
Last edited:
NostaSeronx said:
Oh cool 😀

XOP should technically replace x87 as it can do trignometry and exponential in 128bit and 256bit(I think might be just AVX)
Click to expand...

Although those are probably packed FP format so unless you really wanted all X-bits of precision offered by the x87 internal format, you're kind of screwed.
 
RussianSensation said:
If Person 1 runs 100 meter dash in 10 seconds and Person 2 run it in 20 seconds, then:

1) Person 1 runs 100 meter dash in half the time, or completes the task 50% faster than Person 2 (i.e., (10-20)/20*100 = 50% faster).

2) Person 2 runs 100 meter dash in double the time, or completes the same task 100% slower than Person 1 (i.e., (20-10)/10*100% = 100% slower).

No?
Click to expand...

No.

100% slower means not moving at all.

Person one runs twice as fast, takes half the time. (100% faster, or 2X, or 200% of the speed of the second)
 
Since the only strong point AMD is advertising is the OC capability of these chips I'll try and transalte from the same preview what to expect:

5066MHz with 1.475v was the max freq obtained with a minimum of stability.
The scary part is that when running Cinebench at 4.8 GHZ the Ampermeter showed
22.4 A ~268W. Just imagine what 1.55v means at over 5 GHz...
 
Crap Daddy said:
5066MHz with 1.475v was the max freq obtained with a minimum of stability.
The scary part is that when running Cinebench at 4.8 GHZ the Ampermeter showed
22.4 A ~268W. Just imagine what 1.55v means at over 5 GHz...
Click to expand...

your going to get a perfect semi conductor that spins around a center magnet at near light speeds, able to bend break physics and create antigravity from your desk top machine.
 
Troll Trolling said:
No.

100% slower means not moving at all.

Person one runs twice as fast, takes half the time. (100% faster, or 2X, or 200% of the speed of the second)
Click to expand...

It depends on the scale you choose, that is, you can assign 100% to the slower one, and the other will be 200%. or you can assign 100% to the faster one and the other will be 50%. Elementary.
 
Despoiler said:
Umm you realize the N2 is what you breath in atmosphere?
Click to expand...

LN2 expands 700 times when in turns into gas, and easily displaces and lowers O2 concentration in air to the point which is lethal in an enclosed space. Liquified gases are no joke, even if the gas in question in non-toxic or non-flammable.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top