Sign in with
Sign up | Sign in
Your question

Gaming: Q6600 vs FX-8150

Last response: in CPUs
Share
October 13, 2011 3:59:41 PM

Found these charts on a french website. They are by far the best charts I've found to compare multiple generations of CPUs across both vendors.

http://www.hardware.fr/articles/842-20/jeux-3d-crysis-2-arma-ii-oa.html

The most surprising news to me is that my aging Q6600 O/Ced to 3GHz is *on par* with FX-8150 at stock clocks (I equate Q6600 @3 to be = Q9650 on these charts). And the Q6600 still has room to O/C higher...

Man I love my Q6600. I just wish I hadn't gimped it with a nVidia 680i mobo.

More about : gaming q6600 8150

a b à CPUs
October 13, 2011 4:13:00 PM

I think most reviews I've seen of the new Bulldozer state "disappointed" in their conclusions; sometimes the new CPU falls short to even it's predecessor...
a c 120 à CPUs
a b 4 Gaming
October 13, 2011 4:16:53 PM

Be careful with how you read stats for such a comparison. The FX8150 and Q6600 are on completely different levels of performance and any single (few) benchmark can be deceiving.

Regardless, the Q6600 is a great CPU and performs well for all but the most demanding of users.
Related resources
a b à CPUs
October 13, 2011 5:27:01 PM

Motopsychojdn said:
Check the hardware heaven FX8150 review,
its come to light that all the other reviewers used the same presskit mobo, giving bad results, HH used an Asrock board and things look a hell of a lot better for Bulldozer
http://www.tomshardware.co.uk/forum/315775-10-asus-cros...
Moto
VERY interesting read, thanks
a c 103 à CPUs
October 13, 2011 5:29:56 PM

I was initially gutted (as a lot were) to see 'bad' results on bd,

I've actually bought a 975be as a tide-over chip because I was so disappointed in Bd's initial performance and I've been held up on this build long enough but of course then I saw that thread and maybe, just maybe its still a contender hehe
(pointless now as spent money on the other chip :p )
Moto
October 13, 2011 5:46:13 PM

Motopsychojdn said:
Check the hardware heaven FX8150 review,
its come to light that all the other reviewers used the same presskit mobo, giving bad results, HH used an Asrock board and things look a hell of a lot better for Bulldozer
http://www.tomshardware.co.uk/forum/315775-10-asus-cros...
Moto


Wow those are some quite interesting results that contradict many of the reviews I've seen.

But I would like to criticize the gaming results:
1) They only compared two CPUs. We get no results from Phenom IIs.
2) They only tested at one resolution & graphics setting per game.
3) The resolution (1080p) and graphics settings were set to maximum detail (varying AA). These are situations where performance is more likely to be graphics limited than CPU limited. (okay, looking at Tom's review they did the same maximum detail on all the games).
4) The card they tested with was a 6950. Other sites tested with a 580 because it was top of the line nVidia (before dual GPUs). These guys could have at least chosen the 6970; or even a 6990.

The OC results show a core voltage of 1.5V. Now aren't these chips manufactured on the same processes as Intel's? Because the Tom's article from today (overclocking an i7-2600k) states they are wary of anything above 1.4V and recommend 1.35V for everyday use. Wouldn't 1.5V fry the chip pretty quickly?

I would figure the only difference the mobo makes is with overclocking.

The only legitimate piece of information I get from the review is that there may be some optimizations in the chip/chipset that favor AMD graphics cards by quite a bit. (Good Article Suggestion?)
a c 103 à CPUs
October 13, 2011 6:12:14 PM

I'm taking it as a variation from the 'DOOM!' that other reviews have shown,
I suggested in another thread that Toms can send their chip to members to test on their rigs, results/benches to be posted ofc then the chip returned/ passed onto the next member
that way we see a 'realworld' picture of what the chip does on real builds, rather than accept the inital results and write it off, we can at least see there is hope,
possibly the Mobo used in the kit has some (previously unseen) unknown handicap

I'm not sure about the **The only legitimate piece of information I get from the review is that there may be some optimizations in the chip/chipset that favor AMD graphics cards by quite a bit.**

rather I take from that the fact that the nvidia coding preference (or alleged preference) is taken out of the equasion
I know some games 'prefer' Nvid over AMD cards, you have to rule out all possible variables in my opinion to try and get a true picture
**Yes, the voltage thing is a concern hehe, maybe thats an issue they will address in bios fixes?
Moto
October 13, 2011 6:51:53 PM

Motopsychojdn said:
I'm taking it as a variation from the 'DOOM!' that other reviews have shown,
I suggested in another thread that Toms can send their chip to members to test on their rigs, results/benches to be posted ofc then the chip returned/ passed onto the next member
that way we see a 'realworld' picture of what the chip does on real builds, rather than accept the inital results and write it off, we can at least see there is hope,
possibly the Mobo used in the kit has some (previously unseen) unknown handicap

I'm not sure about the **The only legitimate piece of information I get from the review is that there may be some optimizations in the chip/chipset that favor AMD graphics cards by quite a bit.**

rather I take from that the fact that the nvidia coding preference (or alleged preference) is taken out of the equasion
I know some games 'prefer' Nvid over AMD cards, you have to rule out all possible variables in my opinion to try and get a true picture
**Yes, the voltage thing is a concern hehe, maybe thats an issue they will address in bios fixes?
Moto



Wow, I feel a little redundant in repeating the other forum post you linked to originally.

2 points.

1) This review shows an HD 6970 used as the graphics card. This seems to remove any intel/nvidia conspiracy theories.
http://www.xbitlabs.com/articles/cpu/display/amd-fx-815...

2) This review illustrates how the FX-8150 can be easily CPU-bound where the i7s continue to "stretch their legs" as resolution decreases.
http://www.guru3d.com/article/amd-fx-8150-processor-rev...

Note that in the guru3d article, you could compare Crysis 2 performance at 1920x1080 & 1600x1200 and the framerates are within 2. If you excluded the dataset to this it looks like FX is an awesome processor. But you only get the full story when even lower resolutions are included showing that i7 has quite a bit more "gaming potential". Many of the articles that show the FX & i7 on par with each other do not test resolutions below 1600x1200.

The trend I see is that reviews showing the FX-8150 matching or beating i7-2600k are usually very limited in the amount of data shown & the performance is on par (not FX beating i7 by a large margin). I'm betting they're mostly situations where the system is GPU-bound, but due to the lack of data from lower resolutions we can't prove this.
a b à CPUs
a b 4 Gaming
October 13, 2011 6:58:00 PM

After reading the review I was not really sure why everyone considered it a flop. I was thinking it did quite well in the productivity suites, as for gaming when using gpu dependant games, Why not some RTS games LOADED TO THE HILT with units ?
I Thought things went well during 3ds max and other math based testing.
October 13, 2011 7:07:16 PM

Motopsychojdn said:
Check the hardware heaven FX8150 review,
its come to light that all the other reviewers used the same presskit mobo, giving bad results, HH used an Asrock board and things look a hell of a lot better for Bulldozer
http://www.tomshardware.co.uk/forum/315775-10-asus-cros...
Moto


That's just not true though......i have seen sites using Gigabyte motherboards with 6970 and still see the BD chips fall behind SB.
a c 103 à CPUs
October 13, 2011 7:10:49 PM

Not calling things one way or T'other to be honest,
just saying its not as bad as first impressions might have appeared :) 
Moto
October 13, 2011 7:33:52 PM

Katsushiro said:
Found these charts on a french website. They are by far the best charts I've found to compare multiple generations of CPUs across both vendors.

http://www.hardware.fr/articles/842-20/jeux-3d-crysis-2-arma-ii-oa.html

The most surprising news to me is that my aging Q6600 O/Ced to 3GHz is *on par* with FX-8150 at stock clocks (I equate Q6600 @3 to be = Q9650 on these charts). And the Q6600 still has room to O/C higher...

Man I love my Q6600. I just wish I hadn't gimped it with a nVidia 680i mobo.


The 8150 is actually a great cpu -- misleading reviews are souring peoples mouths.

http://www.tomshardware.com/forum/315805-28-bought-bull...
October 13, 2011 8:22:56 PM

I also have a Q6600 with a slight OC (actually a Q6700, but basically the same thing as the OP's chip). So, I was waiting on BD as a potential upgrade choice. I was really hoping that performance in gaming would be similar to a 2500K/2600K, or that the chip would be competitive enough to at least drive prices down a bit.

Yeah. It didn't happen. The 8150 seems like a complete flop... loses in most highly threaded apps to a 2600K, loses in everything else to a 2500K, only pulls a few slight wins in highly threaded apps, but not enough to chose it over a 2600K in an inexpensive workstation... and the price was too high. I HATE the 8150.
- As a side on that, I really hope that there's some kind of problem with maybe a BIOS version, or OS patch, or just anything, that might help the 8150's standing... but, I'm not holding my breath. And if nothing changes, I think it just SUCKS.

But, from what I can tell, the 4170 seems legit. Good even. It's a bitch finding reviews on the FX-6, and FX-4 chips, due to everyone and their mother looking at the 8150, but my initial info I'm finding on the FX-4 and FX-6 chips seems OK. The per core performance seems about the same as a Phenom II, maybe a little worse, or better, given a situation... power consumption is about the same.... but they OC very well, and the 4170 comes clocked pretty high.

The Phenom II, in my opinion, was the king of the entry level, cheap, gaming processors. And the 4170, actually might fill in pretty well at the low end too. As a solid "good enough" processor to run alongside, say, a 6790 or something, and do some inexpensive gaming on.

So there. I'm trying to be positive. *looks for more 4170 reviews*
October 13, 2011 9:45:16 PM

The only problem with the Q6600 might be with higher end video cards. Could be a bottleneck, more so with SLI. Generally for gaming you are only limited by the GPU, so it does not surprise me that a overclocked Q6600 would preform well VS any current CPU.
October 14, 2011 1:57:52 PM

jonnyrb said:
The 8150 is actually a great cpu -- misleading reviews are souring peoples mouths.

http://www.tomshardware.com/forum/315805-28-bought-bull...


All of the graphs attached to the first post in that thread are misleading. Each one compares performance at a single resolution, which prevents proper interpretation of CPU vs GPU bounding situations.

The reviews I've seen that show multiple resolutions (especially the ones that go down to 1280x1024 or lower) show that the gaming prowess is lacking.

However, in the big scheme of things I will concede that it seems the FX-8150 can handle today's games at very playable settings. I highly doubt that anyone who buys this processor will be disappointed any time soon. Yet as games progress and when Phenom IIs become inadequate, I feel that the FX-8150 will be too. This is one of my complaints: there is little future-proofing value vs high end Phenom IIs.

One more thing, they more than doubled the transistor count vs Phenom IIs and the FX is only marginally faster (Anand states FX is ~2 billion transistors, PII 6 cores are 904 million, PII 4 cores are 758 million). I mean, come on man. This is like if nVidia's Fermi (GTX 580, 3 billion transistors) only beat the GTX 285 (1.4 billion transistors) by like 5-10%. Could you imagine the disappointment if that were the case?
October 14, 2011 3:31:52 PM

Katsushiro said:
All of the graphs attached to the first post in that thread are misleading. Each one compares performance at a single resolution, which prevents proper interpretation of CPU vs GPU bounding situations.

The reviews I've seen that show multiple resolutions (especially the ones that go down to 1280x1024 or lower) show that the gaming prowess is lacking.

However, in the big scheme of things I will concede that it seems the FX-8150 can handle today's games at very playable settings. I highly doubt that anyone who buys this processor will be disappointed any time soon. Yet as games progress and when Phenom IIs become inadequate, I feel that the FX-8150 will be too. This is one of my complaints: there is little future-proofing value vs high end Phenom IIs.

One more thing, they more than doubled the transistor count vs Phenom IIs and the FX is only marginally faster (Anand states FX is ~2 billion transistors, PII 6 cores are 904 million, PII 4 cores are 758 million). I mean, come on man. This is like if nVidia's Fermi (GTX 580, 3 billion transistors) only beat the GTX 285 (1.4 billion transistors) by like 5-10%. Could you imagine the disappointment if that were the case?

Great is an over statement :)  It's more so the other bulldozer are a bit better priced for mid range. the 8150 obviously leaves you wanting a little more.

I love the 580, a very very nice single card, It would be tragic if it would was 5% better than a 285
October 14, 2011 5:42:21 PM

Katsushiro said:
Found these charts on a french website. They are by far the best charts I've found to compare multiple generations of CPUs across both vendors.

http://www.hardware.fr/articles/842-20/jeux-3d-crysis-2-arma-ii-oa.html

The most surprising news to me is that my aging Q6600 O/Ced to 3GHz is *on par* with FX-8150 at stock clocks (I equate Q6600 @3 to be = Q9650 on these charts). And the Q6600 still has room to O/C higher...

Man I love my Q6600. I just wish I hadn't gimped it with a nVidia 680i mobo.

the q6600 is still a decent cpu for most people and unless you expect your computer to be driving tri-sli 590's and being competitive with the i7's.
anyway im not so sure about the q6600@3ghz = to a q9650 tho, as penryn has a 10is% ipc and more cache so it should be about a 3.3-3.4ghz conroe, but then again, my old pc(my mam has it now) has a q6600 in and its being oc'ed to 3.6 with a 400bclk no problems, you should give that a go
!