• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Is The Phenom 1 Really Crap At Games?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

AngelfireUk83

Member
Joined
Nov 7, 2004
I was recently in my local supermarket and I picked up a well known PC mag called MicroMart now current issue has a complete guide to every processor currently out including i7.

I straight away went to the Phenom 1 page and pros was Great at Video Encoding but cons where poor gaming CPU. I agree with video encoding I use Nero Vision from Nero 7 and it vastly improved the times when writing my own created DVD to the HDD. My old 939 3800 X2 took 30-40 mins the 9550 takes 20mins so big improvement, I probably am not even scratching the surface with video encoding and I am looking for some good software (so any suggestions seriously welcome).

But gaming I kinda of disagree I am going to upgrading to 4GB Kingston Hyper X PC8500 next week, New gcard finally at the end of month 4780 whoot and next month 2x Seagate 250gb In RAID 0.

But is the AMD Phenom that crap at games it's long write suggested it was poor framerates at certain res' etc it even blasted the Phenom II in games too.
 
If they were comparing the Phen 1 to newer Intel chips, then I can see why they would say its not great for gaming.
 
The problems with the Phenom1's were heat, clockspeed, and cache. The way I see it, if you are happy with your performance, there is no need to mess with it.
 
Phenom 1 is not better than Athlon 64 X2 in gaming. The X3 series even worse. If you said that they are Crap in Gaming. I am agree with you. Especially if you compared them with Intel CPU at that time. Not just the newer one. Even Phenom 1 isn't as good as Conroe with 4M cache in gaming.

I also agree with Dapman02. It's probably because of it's low clock speed, heat, and cache. And maybe its architecture also cause it.
 
Phenom 1 is not better than Athlon 64 X2 in gaming. The X3 series even worse. If you said that they are Crap in Gaming. I am agree with you. Especially if you compared them with Intel CPU at that time. Not just the newer one. Even Phenom 1 isn't as good as Conroe with 4M cache in gaming.

I also agree with Dapman02. It's probably because of it's low clock speed, heat, and cache. And maybe its architecture also cause it.
Phenom I architecture is NOT a down-grade from K8. That's got to be the silliest thing I've heard in quite awhile.

The clock (or rather over-clock) speeds are the biggest hindrance of Phenom I. But even at lower speeds the Phenom I can do better than an X2 in some games ...
 
Phenom 1 is not better than Athlon 64 X2 in gaming. The X3 series even worse. If you said that they are Crap in Gaming. I am agree with you. Especially if you compared them with Intel CPU at that time. Not just the newer one. Even Phenom 1 isn't as good as Conroe with 4M cache in gaming.

I also agree with Dapman02. It's probably because of it's low clock speed, heat, and cache. And maybe its architecture also cause it.

Plz READ:

http://www.ocforums.com/showthread.php?t=593749
Only because it is the most comprehensive source on the Phenom x2 I have seen and do realize it only gets better if you add cores.

this shows the clear architectural advantages to the L3 cache.
 
When was this guide put out? They could have been comparing it relatively to the rest of Amd's line, which includes the PII and the 940 will beat out any 9950 at stock speeds any day.
 
@base clocks that is not an impossibility as some games love CPU cycles while a lot of everything else uses the entire cpu.

As to the scores posted they (the ones that it lost in) were games that were not taking advantage of the arch improvements. The article points this out. If you look at per core scaling realizing that all 3 cores are not used you will see that the phenom is faster per cycle than the Athlon X2 and that is why there is a BE pay to play.
 
Last edited:
Okay sorry.
I am not saying that Phenom I is a downgrade. But still, it's just no better than Athlon X2 in gaming. And I didn't said Phenom is I worse than Athlon X2. This is based on this review from xbitlabs.
Here's the link : http://www.xbitlabs.com/articles/cpu/display/phenom-x3-8750_6.html
Phenom X3 and X4 lose to Athlon X2 6400 in some games.
And compared to Intel CPU, Phenom I is a crap.
Let's see:

9550 @ 2.2 GHz = 112 Quake 4; 124 Half-Life 2; 18.5 Crysis
6400 @ 3.2 GHz (+45%) = 123 Quake 4 (+10%); 143 Half-Life 2 (+15%); 22.1 Crysis (+19%)
E7200 @ 2.53 GHz (+15%) = 124 Quake 4 (+7%); 144 Half-Life 2 (+16%); 23.9 Crysis (+29%)


Well, the similarly-priced C2D finally shines a little with Crysis but as for the rest? :shrug:
What an interesting interpretation of "crap" ... :)
 
Alright, here is the breakdown of the Phenom versus the other CPU's out there. At the time the Phenom was supposed to take on the Core 2 Quad and the upcoming Penryn. When compared to the Phenom II and the icore series that came out from Intel it could be considered crap. There were and still are several problems with the Phenom. The original phenom without the B3 Stepping I think had a TLB errata problem. They were hot, didn't overclock very well and ate power like a fat boy goes through donuts at Krispy Kreme.

The performance on the X3's was atrocious if I remember correctly. If I recall correctly the X2's would exchange blows and a lot of times the X2's would come out ahead several times. Many reviewer sites did a comparison on the bang for buck Quad Core versus Dual Core a little while after the Phenom came out. Guru 3D came out with the conclusion that bang for buck wise a fast dual core would service you better than a quad core. This is related to the Phenom because a lot performance could be had out of the Phenom, but no where near the crowning performance Intel had.

All in all the Phenom I was not an amazing step forward for AMD. But it was a step forward. Look on any site and you will find the truth about these Phenom I performance numbers. If you want look into the iCore, Phenom II, and Intel's penryn quads and you will see how bad the performance difference is.
 
The trouble with those "bang-for-buck" reviews is they're, literally, old news. When you take into account the current prices of the Phenom I that old picture changes radically. Phenom I's don't run $200+ like they did back then. Now we're down to $120 or less for a 9550 and at the top end $150 for a 9950BE. Go back and check out those bang-for-buck reviews and drop the Phenom I price by 40-60% then see how the numbers fall.


As for comparing a Phenom I to an i7 that's nonsense. I could just as easily show the i7 is junk if I compared it to something costing 3 times as much ...
 
No doubt - but 9550's can be OC'ed as well, albeit usually not as high. I'm not even sure we're talking about OC'ed systems, it's not given one way or another. Since the links are to stock tests that what I compared. IDK (or really care) whether Intel CPU's scale at near 1:1 but several tests I've seen show the Phenom's scaling AT near 1:1.


But, as you well know and have pointed out a few times, even a stock a 9550 is good enough for most games. A highly OC'ed i7 is major overkill for most systems as far as real world playability goes ...
 
^^ just like buying a 295 for gaming. No one REALLY needs it, but most of the time they get it for bragging rights or for F@H and seti, which those I can understand, but bragging rights? Wtf who spends $600 on bragging rights? Why not get a playable system and get good at a game? Then show off your gaming skills instead.
 
well that depends on resolution, i think that if you're at say 1920x1200 1 295 is great or a 4870x2 (I have the later) with AA and such. But when people start putting 2, 3 or 4 of them together thats when it's just like "eh?"
 
Set everything to the same clock speeds and number of cores if you really want to see if there is a performance delta between architectures.

Phenom II is the same breed as Phenom I with possibly 2% increase in performance per clock because of more cache.

There was a review done to see the clock for clock differences between Agena, Deneb, Kentsfield, Yorkfield, and Bloomfield (HT on and off) and it turns out Deneb is still about 4% slower than Kentsfield on average per clock.
 
So what you're saying is - for games there's really no difference ...

For all the games I've played on my Phenom X4 9550 @ 2.2 Ghz I've never had framerate issues.

At increasing resolutions the CPU performance means less and less.

Sure at really low resolutions Intel CPUs win, but who really plays games like that?
 
Back