Sign in with
Sign up | Sign in
Your question

Phenom 1 vs Phenom 2

Tags:
  • CPUs
  • Chip
  • Phenom
Last response: in CPUs
Share
January 28, 2009 3:38:14 PM



Now i'm not saying that phenom 2 is not a good chip, it is only
because of the overclockabilty, when both are run at the same
speeds these two chips are more identical then not.

More about : phenom phenom

January 28, 2009 3:41:59 PM

This is why I say the 9950BE is a better buy than AM2 Phenom 2. Besides, you can get a good 3.2ghz out of the 9950BE which is adequate.
January 28, 2009 3:45:28 PM

you are definitely correct.
Related resources
January 28, 2009 4:06:43 PM

Clock for clock, P2 is wayy faster than P1 is in gaming, and at 3Ghz, P2 uses 80 watts less than P1
January 28, 2009 4:11:13 PM

so really..in a round about way you are saying the Phenom 1's were pretty even with the Kentsfields? then add in the overclockability so they are a better choice all the way around? *shrugs*
January 28, 2009 4:14:06 PM

Kentsfield crushes Phenom 1, its no contest. But when considering price and the performance difference, if I was on a budget I would rather buy the 9950BE.
January 28, 2009 4:14:38 PM

Actualy the phenom 2 are not way faster in games.
January 28, 2009 4:18:12 PM

Yes, and I have seen enough proof that P2 is clearly faster. And runs cooler, and uses alot alot less power. It runs cooler than a Q6600, and it ocees higher as well. So, in essence, P1 is as fast as Kentsfield clock for clock, as youre saying then
January 28, 2009 4:20:52 PM

Just at 3Ghz alone, with the power differences and temps, its clear theyre no wheres near the same
January 28, 2009 4:20:57 PM

JDJ, P2 does not overclock higher than a Q6600. Do not use those LN2 results as backup for this. The Q6600 is good for at least 3.4ghz stable, that's a 1ghz OC, most of them can hit 3.6ghz with well enough cooling. The P2 940 has seemingly struggled to get to 4ghz (a 1ghz OC) on air, let alone 4.2ghz.

There is already a 24 page thread on this so lets not revive it. P2 uses less power and runs cooler than the Q6600 because A. P2 is 45nm, and B. Q6600 is an archaic 65nm design. Even so, only a moron would pick a Q6600 over the P2 940, ill be ditching this in a few days once my Xeon X3370 gets here :D 
January 28, 2009 4:30:12 PM

It ends up higher, tho it also statrts higher. You can get 3.7 on P2, whereas youre lucky to get that at all on a Q660 on air, and P2 can go higher, and hasnt enjoyed the newer steppings as as you said that archaic 6600 has already had.
Remember AMDs slides? They showed around 3.9 was the cut off, and they werent hyping it. Do you remember which slide Im referring to?
"Wow ,, that explains a lot on the why it overclocks so good from a voltage point.

Less TDP = Less Heat = More Oc head room = AMD done us good this time... "

"It is a combo of a lot of things.... this is what AMD should have launched a year ago. This is a great CPU."
Regardless of what some say, others say its a great cpu
http://www.xtremesystems.org/forums/showthread.php?t=21...
January 28, 2009 4:35:33 PM

Yes I remember the slide, the one with the arrows going up the clock speed from air to liquid at 4ghz +. Also, while the P2 940 has a higher stock clock, that has nothing to do with the overclockability.

A Q6600 @ 3.4ghz is the same overclock as a P2 940 @ 4ghz, both are under a 1ghz overclock. Its a 41.6% overclock for the Q6600 and a 33.5% overclock for the P2 940.
January 28, 2009 4:36:50 PM

spathotan said:
JDJ, P2 does not overclock higher than a Q6600. Do not use those LN2 results as backup for this. The Q6600 is good for at least 3.4ghz stable, that's a 1ghz OC, most of them can hit 3.6ghz with well enough cooling. The P2 940 has seemingly struggled to get to 4ghz (a 1ghz OC) on air, let alone 4.2ghz.

There is already a 24 page thread on this so lets not revive it. P2 uses less power and runs cooler than the Q6600 because A. P2 is 45nm, and B. Q6600 is an archaic 65nm design. Even so, only a moron would pick a Q6600 over the P2 940, ill be ditching this in a few days once my Xeon X3370 gets here :D 


If a Q6600 is good for at least 3.4ghz stable, then why is yours only running at 3.2?

The 9950BE would be a good deal if it came down in price a little more. Right now you can get the 140watt version at a decent price but who would want the 140watt version when there is a better and more compatible 125watt version?
January 28, 2009 4:39:45 PM

nsimo86 said:
If a Q6600 is good for at least 3.4ghz stable, then why is yours only running at 3.2?

The 9950BE would be a good deal if it came down in price a little more. Right now you can get the 140watt version at a decent price but who would want the 140watt version when there is a better and more compatible 125watt version?


I keep mine at 3.2ghz because I have a 1.3VID and it runs too hot for my taste. I had it at 3.4ghz for the longest time, but now I keep it at 400x8. I did have it at 3.6ghz but it touches 73c under Prime.
January 28, 2009 4:43:36 PM

You know JDJ the shift in your position is amazing, when phenom
1 came out it didn't overclock well at all, and in all your posts at
that time you wanted the Q6600 to run at the same clock speed
as phenom it was compared to.
Now that phenom 2 can overclock better it's all about the overclock
never mind clock for clock, what happened to the same standards
for both chips.
January 28, 2009 4:52:14 PM

3 Reasons Phenom I is really bad compared to Phenom II

1) At 3Ghz Phenom 1 consumes ~80 Watt more than Phenom II. Which means higher temps and bills. And also shorter motherboard life. The lower power consumption of Phenom II also makes it perfectly compatible with almost all AM2 and AM2+ motherboard, because the actual power consumpion is slightly less than Athlon X2 6000+.

2) Cool N' Quiet has a performance drop of around 30% on Phenom I. None on Phenom II.

3) Both CPU and IMC (NB) overclock good on stock voltage giving greater performance boost.
January 28, 2009 4:58:31 PM

The pnly thing I ever posted about P1 was when I was surprised it was fast enough to bottleneck a few of last gens gpus in games. Other then that, I saw P1 for what it was. I really didnt pay alot of attention to it, but went with mostly what Id heard from around here, and was amazed when it was hyped as being as bad as it was, while a few actuallt said it wasnt that bad, it just wasnt highly clocked.

Dont go around making things up now. Thats the post that caught your attention, and my only post on P1. Ive had more posts on Nehalem and C2D than P1, by huge amounts.

Havinhg P1 running at the same speed as the Q6600? Not sure who youre talking about, but it wasnt me. Ive seen benches with them at stock, and theyre somewhat close, which really means the advantage mainly goes to the Q6600 because of its ocing, with a slight IPC bump
January 28, 2009 5:40:15 PM

I'm not making thing up all i'm asking why all this talk about overclocking
now, when you made it a point to let everyone know that the intel chip
wasn't running at same speeds when AMD is losing in benchmarks.
January 28, 2009 5:56:20 PM

This is what i mean, I said to:


9950 normaly clocked at 2.6 in the article it's overclocked
to 3.2 thats 600mhz overclock if i'm correct.
performed alot worse then all of the intel chips it was up
against. didn't beat or come close to any of them after they
all were overclock it just got worse for the top of the line
amd 9950.

and i said this:

so now it intels fault that the top of the line amd chip
can't overclock higher. well it's also fair to say that before the
overclock the 9950 is 200mhz higher then the q6600 and out
21 test the 9950 won 8 and the q6600 won 13 after the over-
clock not close at all.

Then you said this a few different way about clock speeds JDJ.

I know what youre saying, do you understand what Im saying? They arent at the same clocks, so we dont know what percentage Intel is faster than Phenom, and cant really make a true comparison. Its shows the only ocing on the link Ive provided, that page, and again, like I said, they werent at the same speeds.

and this

Listen real good jed, I know youre a newbie, so listen. Reread what I said about the Intel cpus, just reread it ok? After you read the part about me saying the Intel cpus are clocked higher to their credit, then reread all the times Ive said its too bad they arent clocked the same for a better truer comparison, ok?

and this

Look again, the Intel cpus were clocked higher, to their credit, but its not the same clocks


So why are you talking about overclocking at all when it's all about clock
for clock with you.
a b à CPUs
January 28, 2009 7:55:27 PM

Speaking of overclocking, the i7 ain't so bad as some have stated in other threads: Intel watercools Nehalem to 4.61GHz

Quote:
CES 2009: Francoise's workshop kicks Phenom II back

The guy behind Skulltrail and Skulltrail 2 projects François Piednoel, and the rest of its performance analysis group, have showed us Nehalem, Core i7 PC water cooled all the way to 4.61GHz.

They did it on an Intel desktop motherboard powered by Intel's desktop control center, something that lets you overclock from Windows. The voltage was all the way to 1.44V, whereas the chipset needed 1.21V.

...

The CPU temperature was at 61 degree Celsius while the CPU Voltage regulator was at 31 °C and chipset at 41 °C. The clock is much higher than the one you can score on Phenom II, but Intel’s CPU used here is much more expensive.

Nehelem, Core i7 is definitely better water overclocker than Phenom II but Phenom II still overclocks higher at 6GHz+ domains.


Of course, watercooling is much more practical than LN2 or LH2 :) 
January 28, 2009 9:00:44 PM

but compared to Intel the P2 uses way too much voltage. You're talking 1.55v to get 3.7-3.9Ghz stable? Where the Q9650 you can hit 4GHz on 1.3v
January 28, 2009 9:35:14 PM

blackpanther26 said:
but compared to Intel the P2 uses way too much voltage. You're talking 1.55v to get 3.7-3.9Ghz stable? Where the Q9650 you can hit 4GHz on 1.3v



Whaaa?

Im sitting at 3.7ghz with a little ole 1.4625 vcore! I think you need to do some research before spouting off like that. And before anyone say's its not stable ive been running Xilisoft for the last 13hour's encoding southpark with all 4 cores at 100%.

How much vcore would a q6600 need to reach 3.7ghz?
January 28, 2009 9:44:02 PM

Quote:
Whaaa?

Im sitting at 3.7ghz with a little ole 1.4625 vcore! I think you need to do some research before spouting off like that. And before anyone say's its not stable ive been running Xilisoft for the last 13hour's encoding southpark with all 4 cores at 100%.

How much vcore would a q6600 need to reach 3.7ghz?


A better ? how much vcore would it take to get a phenom 1 to 3.7.
January 28, 2009 9:45:45 PM

My Q6600 needed 1.5v to hit 3.8ghz. 3.6ghz was around 1.46. Temps at 3.8ghz were beyond reason though, it was idling at like 55c.
January 28, 2009 11:17:31 PM

OK
@ jed, All Im saying is, its hard to tell clock for clock to compare P1 to Q6600, as theres many ways to get P1 to oc speeds besides multi, and it effects its performance, tho using a simple multi, then at same clocks, wed know how much faster Kents is than P1 clock for clock

@blackpanther26, you do know that voltage has nothing to do with power usage? P2 starts at 1.35v stock, so % wise,.125v isnt much for a 700Mhz oc

As far as P1 hitting 3.7, it isnt using P2s superior immersion process, which is very similar to HKMG in what it does for temps/power, when AMD does implement HKMG, itll truly be a cool and low powered cpu, better than anything we see now, including Intel, tho Intel will catch up and be using immersion by then as well
January 28, 2009 11:25:46 PM

yea i would imagine immersion plays into the next die shrink.
such a strange thread. i thought it was already well established the p2 was a ton better than p1. is it that the p1 is seen as better than originally thought or is this to dismiss the gains in the p2?
January 28, 2009 11:32:04 PM

Quote:
@ jed, All Im saying is, its hard to tell clock for clock to compare P1 to Q6600, as theres many ways to get P1 to oc speeds besides multi, and it effects its performance, tho using a simple multi, then at same clocks, wed know how much faster Kents is than P1 clock for clock


This holds true for phenom 2 as well run them at the same clock to compare,
all i'm saying is i don't see you mentioning the clock for clock now that phenom 2
is better at overclocking then phenom 1.
January 28, 2009 11:34:31 PM

You see, when I made the thread about P1 being as fast as it was, people jumped all over me, and I wasnt looking thru rose colored glasses at all then or now either, so yea, I think its to get people thinking P1 is similar to P2, when they arent anything alike, for power, performance and temps
January 28, 2009 11:37:59 PM

at the same time, the p1 didn't earn the bad rap it got. it got harsh treatment because it couldn't really overclock enough to be enticing to an enthusiast but most people would never be the wiser. the bad publicity made a decent chip out to be a turd.
January 28, 2009 11:53:03 PM

And it came in at low clocks too, like the Q6600, and THEN couldnt oc, and that made the 6600 look way better .
January 29, 2009 12:04:05 AM

By time this community finally sorts out and agrees to disagree on these chips AM4 will be here.

I would have been more than willing to get that snazzy Gigabyte AM3 790FX and a P2 945, but with the recent price drops the Q9650 (or Xeon X3370 in my case) was the most obvious best buy.
January 29, 2009 12:39:46 AM

of course. you already have everything but the cpu. makes perfect sense to me!
January 29, 2009 1:22:56 AM

xx12amanxx said:
Whaaa?

Im sitting at 3.7ghz with a little ole 1.4625 vcore! I think you need to do some research before spouting off like that. And before anyone say's its not stable ive been running Xilisoft for the last 13hour's encoding southpark with all 4 cores at 100%.

How much vcore would a q6600 need to reach 3.7ghz?



That is still too much voltage speeking the 45nm Core 2 Quad's uses less voltage which means a higher overclock and lower wattage.
January 29, 2009 1:45:51 AM

JDJ what cpu are you using anyway.
January 29, 2009 5:45:43 AM

Amount of voltage is different for each process.

(I make no guarantees on the accuracy of my information, but it's all to the best of my knowledge)

Intel's 45nm HKMG process lowers voltage requirements, but at the same time doesn't handle high voltage near so well as AMD's non-HKMG process.

What's really important isn't 1.5v versus 1.4v, it's the difference. If Intel starts at 1.3v, going to 1.4-1.5v isn't a *huge* difference. If AMD starts at 1.15v and needs the same 1.4-1.5v, that's much worse.

Basically, stop arguing about what's too much voltage across processes/architectures. Each one handles voltage differently. Higher power usage, bad, but argue about power and heat, not the voltage.
January 29, 2009 4:41:57 PM

Sorry I've gotta say it Voltage=Crap, Amperage=DEAD!. (in other words I wish the manufacturers would state the amperage draw of these chips/motherboards). In laymans terms 1,000V@.002Amps=you survive, 1V@25Amps=You die.
!