Anand: PII vs. Q9550 vs. i7 crossfire, Phenom II = smoother

[keithlm]

lol ninja edit to save your face

No not really.

The sad part is that the WORDING you choose often makes you a troll. I had put something about your wording... but decided it wasn't worth bothering.

Just like replying to you is generally never worth the effort.

[iandh]

hmmmm I seem to recall a pretty important person 'round here sigged



Can't for the life of me remember who though

With all due respect to the owner of this forum, and no offense intended, that quote is BS, ignorance, and mob mentality all rolled into one.



I'll put it this way. If you sat me down in front of my Q6600 system and my Phenom 9950 system in a blind test , I could tell you in an instant which was which. There is not a single doubt in my mind.

I would be willing to perform this test and post a video but then a bunch of trolly trolls would come along and say it was staged, so it would be no use.

Either way, please don't come and post an obvious derailment/flame/troll post like this right after a mod posted a warning and we just got the thread back on track into a workable discussion. It does nothing but further prove the point that intel users are interested in nothing else but doing everything they can to ruin every good AMD discussion they come across.

[KELL5]

.Either way, please don't come and post an obvious derailment/flame/troll post like this right after a mod posted a warning and we just got the thread back on track into a workable discussion. It does nothing but further prove the point that intel users are interested in nothing else but doing everything they can to ruin every good AMD discussion they come across.

Not at all, you've taken more from my statement than was required. I've never bashed amd on this forum, nor intel. I ain't a fanboy. I simply made a statement which is true, and in jest at that. No need to run to the rescue. No fire here.

I've always agreed my phenom rig purrs along much nicer than my C2Q when multiple tasks are running. I also remember a few guys being slated here when they said similar. Nothing to worry about. Don't be so touchy maybe.

[Titan7171]

What I find a little perplexing is why if many people have known that Phenom runs very smooth and that now a website confirmed they feel the system feels very smooth........why we need tests again? I have a feeling people want to make it as difficult as possible to accept what many have said without tests and graphs. Lets just drag this out as long as possible.........there is a savior amongst us, he has come from afar with tidings of good will to teach us all.

One of these things is not like the others,
One of these things just doesn't belong,
Can you tell which thing is not like the others
By the time I finish my song?

Did you guess which thing was not like the others?
Did you guess which thing just doesn't belong?
If you guessed this one is not like the others,
Then you're absolutely...right!

[iandh]

Not at all, you've taken more from my statement than was required. I've never bashed amd on this forum, nor intel. I ain't a fanboy. I simply made a statement which is true, and in jest at that. No need to run to the rescue. No fire here.

That's why I tried to keep my post somewhat toned down. I wasn't sure if you were jesting or actually trying to cause a derailment.

The bottom line is that the quote "Phenom + BS = Smoother" may apply very well to a downtrodden AMD fanboy who is sad about his shrunken e-peen, but the quote becomes much harder to validate when "Phenom = Smoother" is uttered by a well known review site that is generally thought of as biased towards Intel.



Either way I know that I use my Phenom to game and my Q6600 for video encoding and crunching based upon my personal experience, and there isnt a single spec of BS involved.

What I find a little perplexing is why if many people have known that Phenom runs very smooth and that now a website confirmed they feel the system feels very smooth........why we need tests again? I have a feeling people want to make it as difficult as possible to accept what many have said without tests and graphs. Lets just drag this out as long as possible.........there is a savior amongst us, he has come from afar with tidings of good will to teach us all.

One of these things is not like the others,
One of these things just doesn't belong,
Can you tell which thing is not like the others
By the time I finish my song?

Did you guess which thing was not like the others?
Did you guess which thing just doesn't belong?
If you guessed this one is not like the others,
Then you're absolutely...right!

It's because some people are so fascinated with "who has the biggest graph" that if they chose to accept that "who has the biggest graph" is not the single binding force in the universe, it would cause their whole world to come crashing down.

[Hornet331]

No not really.

The sad part is that the WORDING you choose often makes you a troll. I had put something about your wording... but decided it wasn't worth bothering.

Just like replying to you is generally never worth the effort.

and your better?

its the first time you actually posted data regarding to this topic (including the thread from january), rather then calling everyone a troll that not agrees with your opinion.

I have my standpoint regarding to this topic, but im not so close minded that i call everyone a troll that has its own standpoint. I dont denie the possibility that phenom could be smoother then C2 or Ci7, but as it stands right now thers no evidence for that conclusion.

If there is sufficent data that proves any of the possible outcomes i accept it as fact. I am curious about that "phenom is smoother" ever since the first topic surfaced back in sept. 08.

[KELL5]

That's why I tried to keep my post somewhat toned down. I wasn't sure if you were jesting or actually trying to cause a derailment.

The bottom line is that the quote "Phenom + BS = Smoother" may apply very well to a downtrodden AMD fanboy who is sad about his shrunken e-peen, but the quote becomes much harder to validate when "Phenom = Smoother" is uttered by a well known review site that is generally thought of as biased towards Intel.



Either way I know that I use my Phenom to game and my Q6600 for video encoding and crunching based upon my personal experience, and there isnt a single spec of BS involved.

Couldn't agree with you more.

AMD FTW

back on topic....I agree the tests here shouldn't be used to "rub it in the face" of intel users as such. I do however find it interesting to try and possibly source the reason for this "smoother" feel.

[Titan7171]

I never saw any data that said Intels offerings were smooth at all. So who says it is?

[iandh]

I never saw any data that said Intels offerings were smooth at all. So who says it is?

Yeah there isn't any data saying intel systems are smooth and I am not going to go on just user opinion alone so they need to PROVE IT. I'm sure their battle will be up at least as steep of a hill as ours.

[roofsniper]

What I find a little perplexing is why if many people have known that Phenom runs very smooth and that now a website confirmed they feel the system feels very smooth........why we need tests again? I have a feeling people want to make it as difficult as possible to accept what many have said without tests and graphs. Lets just drag this out as long as possible.........there is a savior amongst us, he has come from afar with tidings of good will to teach us all.

the problem is that the only data we have to support our claims right now is user opinion. even tho AT said the phenom system looked smoother theres no actually data you can view. you can only go off of what is said. and as you can see below hornet even confirms this. we can say w/e we want but it won't be taken seriously until there is data to be shown.

I dont denie the possibility that phenom could be smoother then C2 or Ci7, but as it stands right now thers no evidence for that conclusion.

If there is sufficent data that proves any of the possible outcomes i accept it as fact. I am curious about that "phenom is smoother" ever since the first topic surfaced back in sept. 08.

[Titan7171]

Yeah there isn't any data saying intel systems are smooth and I am not going to go on just user opinion alone so they need to PROVE IT. I'm sure their battle will be up at least as steep of a hill as ours.

thats exactly what Im saying........they need to figure out why AT found it smoother...why the I7 and Q wasnt as smooth... we dont have to find out anything,... its smooth already why should I question what I know?

[roofsniper]

Yeah there isn't any data saying intel systems are smooth and I am not going to go on just user opinion alone so they need to PROVE IT. I'm sure their battle will be up at least as steep of a hill as ours.

the argument for saying intel systems are smooth is the fact that they get a higher average fps. and if both systems get the same average fps then they are equally smooth. the problem is that this isn't 100% true because you need to look at the big picture which this testing should show.

[Titan7171]

the problem is that the only data we have to support our claims right now is user opinion. even tho AT said the phenom system looked smoother theres no actually data you can view. you can only go off of what is said. and as you can see below hornet even confirms this. we can say w/e we want but it won't be taken seriously until there is data to be shown.

And I have to prove myself to whom?? tell me who the hell is anyone to question and make you prove anything.........let them find out on their own........Im not jumping through hoops for no ing body

not trying to dis you but Im just saying who cares what they think? lol

Just had to add we bought our cpu's before this tests was done by AT so it changes nothing for us.

[gosh]

I think that the focus on smoothnes should include the mouse and how the processor handles the unexpected. How responsive is the mouse on each processor if the processor works hard or if there is very much data that is sent to or from the cpu to other hardware. What happens if one thread is moved from one core to another core etc. If the game behaves as the developers has planned, it may be harder to understand how the cpu behaves.
Personally I think it is a communication thing. The mouse is an I/O operation and I/O operations need to complete before the CPU switches to another thread on that core if it is not made asynchronously. Threads need to be balanced if the cpu should be used effectively. If one thread does more work compared to another and they both need to synchronize, then this will delay both. If something happens in one thread and it slows down, this will slow down other threads until the unexpected has passed away.
Heavy traffic between the CPU and other hardware and the unexpected is more likely to show up.
If you have 200 FPS but the mouse readings are bad, the game will not feel smooth.

[roofsniper]

responsiveness and smoothness in games are two different things tho.

[iandh]

the argument for saying intel systems are smooth is the fact that they get a higher average fps. and if both systems get the same average fps then they are equally smooth. the problem is that this isn't 100% true because you need to look at the big picture which this testing should show.

The funny thing is one would think that they all would be falling all over themselves making excuses for why the intel rig stuttered in Crysis, a game that supposedly "plays best on intel", but instead they bash the clean running AMD rig in an attempt to draw attention away from the fact that the Q9550 sucked on that title.


I should edit my sig to my intel rig, and go over to the intel forum and post "Anand: Q9550 stutters in Crysis" (but not mention anything at all about AMD or PII), and then see what hilarity ensues.


I don't actually have the balls to do so though, it would probably irritate XS staff just a hair too much by doing that.

I think that the focus on smoothnes should include the mouse and how the processor handles the unexpected. How responsive is the mouse on each processor if the processor works hard or if there is very much data that is sent to or from the cpu to other hardware. What happens if one thread is moved from one core to another core etc. If the game behaves as the developers has planned, it may be harder to understand how the cpu behaves.
Personally I think it is a communication thing. The mouse is an I/O operation and I/O operations need to complete before the CPU switches to another thread on that core if it is not made asynchronously. Threads need to be balanced if the cpu should be used effectively. If one thread does more work compared to another and they both need to synchronize, then this will delay both. If something happens in one thread and it slows down, this will slow down other threads until the unexpected has passed away.
Heavy traffic between the CPU and other hardware and the unexpected is more likely to show up.
If you have 200 FPS but the mouse readings are bad, the game will not feel smooth.

Agreed. Come to think of it, the stuttering is more apparent on intel rigs for me when mousing, and not when strafing. In fact, when playing Crysis on my Q6600 rig I found myself using strafing more often as a crutch in aiming. Typically mousing requires more of a scene change than strafing, so perhaps that is a clue... dunno.

[Mav451]

Ugh...please stop with the partisan b.s.

I'm patiently waiting for hard numbers that validate/refute the smooth claim. On the same token, there are also no numbers that validate/refute the notion that Intel stutters. Let the numbers speak for themselves.

[roofsniper]

Ugh...please stop with the partisan b.s.

I'm patiently waiting for hard numbers that validate/refute the smooth claim. On the same token, there are also no numbers that validate/refute the notion that Intel stutters. Let the numbers speak for themselves.

well we need a tester and a game/testing method. as hornet said we could ask JJ. im not sure who else could do it.

[iandh]

Ugh...please stop with the partisan b.s.

I'm patiently waiting for hard numbers that validate/refute the smooth claim. On the same token, there are also no numbers that validate/refute the notion that Intel stutters. Let the numbers speak for themselves.

Yeah, I'm incredibly partisan when I have a whole closet of full of both AMD and intel hardware... I'm making fun of folks who are partisan, not being partisan myself.

(unless of course I am partisan by being a member of the non-partisan party)

[keithlm]

and your better?

its the first time you actually posted data regarding to this topic (including the thread from january), rather then calling everyone a troll that not agrees with your opinion.

I have my standpoint regarding to this topic, but im not so close minded that i call everyone a troll that has its own standpoint. I dont denie the possibility that phenom could be smoother then C2 or Ci7, but as it stands right now thers no evidence for that conclusion.

If there is sufficent data that proves any of the possible outcomes i accept it as fact. I am curious about that "phenom is smoother" ever since the first topic surfaced back in sept. 08.

I don't call everybody a troll. Only people that intentionally derail threads. This would include you. Thank you for again confirming my opinion that you are a troll and really not worth bothering to reply to.

YOu can claim to be "open minded" but the problem is that your posts are not conducive to an active conversation about a topic; they make it abundantly clear that you have made up your mind and are not willing to contribute to the conversation in a positive manner. Sadly you don't realize that you can troll without actually having the blatant intent.

[roofsniper]

i don't know why everyone calls hornet a troll can we just drop it and stay on topic. i would be interested to see these results but right now we are going no where. find someone who could do testing for both systems and we need to think of a game/testing method.

[Raiderman]

The only thing I could contribute to this conversation is this...I don't know! Sure is interesting to view everyone's opinions though

[Raiderman]

i don't know why everyone calls hornet a troll can we just drop it and stay on topic. i would be interested to see these results but right now we are going no where. find someone who could do testing for both systems and we need to think of a game/testing method.

Chew* is in the process, as you know, of testing a Q6600, and a Ph II. Maybe we could find a testing method, and ask him to do it.

[Mav451]

You know I just had a thought - if we indeed are testing for smoothness (and not raw performance), then the focus should not be on using the same game/hardware.
Think about it - GPU-intensive game; low-frame rate. Therefore, MORE time between each frame generated. One result.
OTOH, consider low GPU-intensive game; high-frame rate. Time between each frame generated will be considerably smaller. Another result.

Point is that we don't care about the differences, we are ONLY looking for the LACK of consistency in the time between each frame. This is why the focus is on the derivative of the data generated, not on the absolute values.
E.g. the focus is not on why a 60fps bench is different than 30fps bench. The focus is on which has a more WILD delta in the rate of frames generated, relatively. Taking the derivative of the data will make sure the differences are relative to the data sets.

Haha am I making sense here?

[roofsniper]

Chew* is in the process, as you know, of testing a Q6600, and a Ph II, maybe we could find a testing method, and ask him to do it.

yes but he is testing for performance. it is possible that he could change his systems up some to perform the test tho.

You know I just had a thought - if we indeed are testing for smoothness (and not raw performance), then the focus should not be on using the same game/hardware.
Think about it - GPU-intensive game; low-frame rate. Therefore, MORE time between each frame generated. One result.
OTOH, consider low GPU-intensive game; high-frame rate. Time between each frame generated will be considerably smaller. Another result.

Point is that we don't care about the differences, we are ONLY looking for the LACK of consistency in the time between each frame. This is why the focus is on the derivative of the data generated, not on the absolute values.
E.g. the focus is not on why a 60fps bench is different than 30fps bench. The focus is on which has a more WILD delta in the rate of frames generated, relatively. Taking the derivative of the data will make sure the differences are relative to the data sets.

Haha am I making sense here?

we need a core 2 system that gets the same average fps as a phenom system. so each system would have the same performance. but then if you look at the chart and you can see the time it takes for each frame to be produced it will say more. in this situation a straight line would be the best results. showing that it took the same amount of time for every frame. so we are looking for the straightest possible line. a jagged line would show that the frames are coming out at random times instead of at a smooth rate.