AMD Phenom to Phenom II Upgrade Article from Corsair

Gah. Benches are basically useless because too much of the system changed between them. In general, they are only useful if one part of the system changes, not just about everything. Also, you used AS5 and a decent aftermarket cooler and didn't overclock...what was the point of the aftermarket cooler? Might as well have used the stock cooler.
Also, 3dmark06 is largely a useless bench. Just bench using otherstuff like the crysis benchmark tool.
 
Gah. Benches are basically useless because too much of the system changed between them..[/URL]

That's because this is an upgrade article. I suppose I could have benched each CPU head to head with all possible configurations of GPUs, and on both OSs, and then did the same with the HDs, etc etc but that is WAY beyond the scope of the article.

And, I used some real world tests to show what the actual gains were. A Crysis bench using an HD3870 is pretty much worthless.
 
Let me rephrase what I wanted to say a bit better:

1. Overclock the Phenoms.
2. You should have just left out the 3870 and used a 4870 in both systems. Everyone knows how crappy a 3870 is. You don't really need to show this.

Right now here is what people want to know:
I love gaymen should I upgrade my CPU?
Should I upgrade my harddrive?
What are the advantages of using an SSD drive?
 
Last edited:
What you are describing is outside the scope of the upgrade. It's easy to forget that a majority of the world's computer users do not upgrade as often as the average [H] reader. So, it's realistic to consider that they may have older CPUs, GPUs, and HDs than many people here. There are more people in the world using CPUs at stock speeds than there are OCers. So, both CPUs head to head at stock speeds keeps it realistic for a larger number of people.

I also don't have an X1900 series GPU to compare so I used the oldest card I have access to. The results show a great contrast and what can be gained for a $150 GPU upgrade.

It seems that you are not a person that is going to benefit from what is in this article which is fine.But, the results and methodology are still valid.
 
I clicked expecting the only change to be the cpu to be the change.

I don't really see the point in swapping out so many components.
 
I clicked expecting the only change to be the cpu to be the change.

I don't really see the point in swapping out so many components.

It's a $500 upgrade and it's only 3 components.

Multitasking kills the Phenom 9500, the 940 is a great upgrade.

Any good game is going to bog the HD3870. The HD4870 is a great upgrade.

People want to see what SSDs are all about. The numbers speak for themselves.

So, for $500 we've made significant improvement over a wide variety of uses.
 
Who exactly are you writing your article for, who is reading your article, and what are you trying to accomplish with this article?
I can't really answer these questions. But I can answer this though: for the people who are interested in hardware this article falls short. The results that are only affected by one component are valid but the rest don't really say anything. Also, you could have said something more about the SSD. 20s off OS load time doesn't justify the cost of an SSD. I get a bunch of numbers but how do they translate into performance? This should have at least been explained.
 
maybe swap one thing out, bench it

swap the next thing out, bench it

swap the next thing out, bench it.

like a stair stepped approach

then the whole thing.

I would like to know how much each component is worth, I can't afford 500 bux, but I can afford 150 here 200 there 150 there.
 
maybe swap one thing out, bench it

swap the next thing out, bench it

swap the next thing out, bench it.

like a stair stepped approach

then the whole thing.

I would like to know how much each component is worth, I can't afford 500 bux, but I can afford 150 here 200 there 150 there.
It is only needed for some tests to swap stuff out. Also, I agree with the 500$ statement.
 
I didn't realize you could fit all that in for $500, that's pretty fucking cool. We already knew the Phenom II was a winner, though :)



Oh sorry, forgot I was supposed to whine like a little pussycake because the review didn't go the way I expected.
 
jerks who talk shit just to talk shit to complete strangers on the internet are really cool.
 
Oh sorry, forgot I was supposed to whine like a little pussycake because the review didn't go the way I expected.
This is called giving feedback. Do you think he wants to hear "WOW THIS IS A GREAT ARTICLE!!" from everyone? I do not think he came here looking to increase his ego. If you have the time, giving feedback is a nice thing to do.
 
Joseph, he already stated his reasoning for doing the article in the way he did. There is no need to constantly go at him question after question, he stated his intent and if you don't like it, that's fine, just move on.
 
good article, im sure there are quite a few people, myself included, that upgrade small pieces here and there... its an good way to show how to upgrade for a good bang for your buck... the only complaint is that some of the screen shots need to be bigger so i can see the numbers
 
why doesn't 3dmark accurately describe how my system will perform in every game :(
 
why doesn't 3dmark accurately describe how my system will perform in every game :(


because its cpu bound and not gpu bound.. thus if you have something like an i7 vs phenom II 940.. the i7 will murder it in 3dmark06 running the same card.. yet in almost all games they run the exact same frame rates with the exact same settings.. 3dmark has gone to absolute shit over the years.. and the crysis benchmark tool is really only good to tweak settings for the game and check stability.. it really doesnt say much about the system.. since the game barely touches the cpu.. all it really tells you is how good of a gfx card you have.. nothing else..

each benchmark program has its plus's and cons.. though everyone knows about 3dmark06 so that tends to be the benchmark tool used by everyone.. even though most of the enthusiast/overclocking users knows its a pos..

personally i like using cinbench10 to benchmark my cpu..

-----------------------------------------------
after reading the article.. i have to agree the pictures are way to small.. would be nice if you could click on them to make them larger or something.. on a 24" monitor its just impossible to read..

as for the article its self.. its not bad.. it definitely is geared more toward the under educated(cant think of anything else to call it) user.. and not the enthusiast users that frequent [H]..

also id have to agree it would be cool just to see some basic overclocking numbers.. nothing insane.. maybe something like 3.6ghz.. it would also help get the point across in why black edition processors are worth the premium over say the phenom II 920..

but overall i think the articles fine..

oh yeah.. another thing.. ya might want to add temps when you talk about the Scythe Mugen 2 to show why aftermarket cooling is a good idea over the stock cooling at idle and full load..
 
Last edited:
jerks who talk shit just to talk shit to complete strangers on the internet are really cool.

I apologize for my lack of civility. I realize now that I should be more respectful towards girls.

This is called giving feedback. Do you think he wants to hear "WOW THIS IS A GREAT ARTICLE!!" from everyone? I do not think he came here looking to increase his ego. If you have the time, giving feedback is a nice thing to do.

The article wasn't written with [H] readers in mind. He's explained this.
 
because its cpu bound and not gpu bound.. thus if you have something like an i7 vs phenom II 940.. the i7 will murder it in 3dmark06 running the same card.. yet in almost all games they run the exact same frame rates with the exact same settings.. 3dmark has gone to absolute shit over the years..

Partially true. You do get an increase in sm 2.0 & 3.0 scores with a faster cpu, but not the large disparity with overall score that having an i7 vs. a Phenom II. If I compare my system to a c2Q rig with the same video card we get similar scores for sm 2.0 & 3.0. Against an i7 rig with an overclocked (800/1100) 9800gt I trail by about 1500 points in sm 2.0 & 3.0 which is to be expected but a whopping 3000 points in cpu score.
 
The lack of scientific methodology could leave beginners lost.

Not only that, the article didn't go into power consumption, total system weight, or heat output, the whole thing is obviously flawed.

internet forums are full of immature morons, as soon as I forget I am reminded.

Try to stay on topic.



Based on my highly scientific analysis of Newegg.com, you could get the CPU and 4870 for $355 + shipping. Personally I would skip the SSD in this scenario.
 
Last edited:
way to thread jack this guys thread with your own stupidity.

you did no good and only wanted the spotlight.

thank god I can put you on ignore and never read your dribble again.

To the OP, sorry, I was just trying to help, the article was good, I just have been contemplating a change to a phenom II and was hoping for something else based on the thread title, I also can't afford 500 bux in a single upgrade.

I have no idea how to use the article to help me, but I am sure it is helpful to others, wasn't trying to turn this thread into what it has turned into.

humble apologies.
 
I found the article useful, I like to upgrade a few pieces at a time being a student and all. Plus I started with the same proc he had, except I went on to AM3 opposed to the top end of AM2+.
 
Not only that, the article didn't go into power consumption, total system weight, or heat output, the whole thing is obviously flawed.



Try to stay on topic.



Based on my highly scientific analysis of Newegg.com, you could get the CPU and 4870 for $355 + shipping. Personally I would skip the SSD in this scenario.


the problem is that the majority of uneducated computer users.. have no clue about power consumption, total system weight or heat output.. and if he was to get into the whole power output per hardware.. then you get all the confused little kiddies asking what the difference is between amd's 125w phenom II vs intel's 130w i7.. and the problem with that is you could write an entire 10 page article describing the difference between amd's wattage rating and intel's wattage rating..

but i will agree it would be nice to see the heat output like i said further up in my last post.. but overall the majority of users wouldnt know the difference between temps.. nor do most of them even understand the difference between celcius and fahrenheit in the US and would lead to even more confusion.. like most of us have been saying.. this article was writen for the casual computer user/gamer.. not for people like us that live and die by real world computer performance..
 
way to thread jack this guys thread with your own stupidity.

you did no good and only wanted the spotlight.

thank god I can put you on ignore and never read your dribble again.

To the OP, sorry, I was just trying to help, the article was good, I just have been contemplating a change to a phenom II and was hoping for something else based on the thread title, I also can't afford 500 bux in a single upgrade.

I have no idea how to use the article to help me, but I am sure it is helpful to others, wasn't trying to turn this thread into what it has turned into.

humble apologies.


not only in the article does he explain that its a 500 dollar upgrade(at the time of writing it) but he also shows how much each part cost in the upgrade.. at the time of him writing that.. the phenom II 940 was 190 dollars.. its on this page.

in all honestly the phenom II 955 is the better buy if you have the money.. but for a 500 dollar budget upgrade the 940 is a very good choice..
 
This review honestly is worthless to me because:

1. There are already performance reviews pitting Phenoms against Phenom II's so we know how much faster the PII is than its predecessor.

2. Everyone already knows that a total system upgrade is going to be faster in EVERYTHING and somehow we made it by all these years without someone stating the obvious.

3. You didn't use the Phenom 9950. If you're going to use the best AM2+ Phenom II then you need to counter with the best Phenom. This is like saying an upgrade from a Q6600 is going to be slower than an i7 965 . . . NO SHIT.

4. You should have indeed run the 3870 and 4870 in both machines at the very least.
 
Last edited:
ok, we get it, this article wasnt about you so youre not happy, lets hae a pity party

This review honestly is worthless to me because:

2. Everyone already knows that a total system upgrade is going to be faster in EVERYTHING and somehow we made it by all these years without someone stating the obvious..

3. You didn't use the Phenom 9950. If you're going to use the best AM2+ Phenom II then you need to counter with the best Phenom. This is like saying an upgrade from a Q6600 is going to be slower than an i7 965

this is the most stupid thing i have ever heard, he didnt make the article to compare the old fastest to the new fastest, and did you even read the article?

again this is for the CASUAL computer user, i know many people that "know" computers but more from software side and know nothing about hardware, and an article like this is perfect for them

again get over yourself, just because the article isnt geared toward you doesnt mean its useless, it just means it written with a different audience in mind
 
Thanks for all the feedback, positive and negative.

One side note I did not include in the article was that I started out with a MOBO from another manufacturer than I won't name. It was an older board more suited for the scope of this article that would run both CPUs. It was also a more logical pairing with and older Phenom. But, .....LOL...the MOBO gave up the magic blue smoke about 1/3 of the way through so I had to start from scratch using a newer board. However, either board was perfectly capable of highlighting what an user with a similar system could expect if they spent $500 on an upgrade as opposed to building a totally new system.

I think one of the best messages coming across through this specific discussion is that this article is obviously not geared towards every computer user. However, that does not diminish its usefulness. As noted, we did this directly in response to consumer demand.
 
Thanks for all the feedback, positive and negative.

One side note I did not include in the article was that I started out with a MOBO from another manufacturer than I won't name. It was an older board more suited for the scope of this article that would run both CPUs. It was also a more logical pairing with and older Phenom. But, .....LOL...the MOBO gave up the magic blue smoke about 1/3 of the way through so I had to start from scratch using a newer board. However, either board was perfectly capable of highlighting what an user with a similar system could expect if they spent $500 on an upgrade as opposed to building a totally new system.

I think one of the best messages coming across through this specific discussion is that this article is obviously not geared towards every computer user. However, that does not diminish its usefulness. As noted, we did this directly in response to consumer demand.

I think it's a very valid article, just probably not here on this forum. Doesn't hurt to share though.

I see these people everytime I go to Microcenter. They are asking questions, like what is outlined in the article, to the moronic sales staff at the store. Sometimes the employees give a good an honest answer, but most of the time they feed them a bunch of BS. Maybe one of those poor saps was suckered into a Corsair memory upgrade that they didn't need and found this article on your website....

Sorry for the ramble.....just got a memory of some guy that was being fed a whole lotta BS out at the store once.
 
I can only imagine how many 10s of thousands of people visit this site each month. And search engine propogation will also lead people here so hopefully even here at the [H] the article will help some people that can benefit from it.
 
Ugh... the parts you start off with in the article are WAYYY out of line for what a noob would even consider to begin with. C'mon a $180 board, a $150-$200 psu, dominator sticks of ram...? Seriously? This is why ppl are saying, lol wut??
 
Ugh... the parts you start off with in the article are WAYYY out of line for what a noob would even consider to begin with. C'mon a $180 board, a $150-$200 psu, dominator sticks of ram...? Seriously? This is why ppl are saying, lol wut??

For whatever reason, your argument of what you think customers want is not in line with what we are seeing at the support level. As noted, we did this in direct response to customer inquiries. Not every customer that approaches us for presales or upgrade advice is crusty old [H] reader.

This is not to say that every single one of them has a Phenom 9500 or the same Maxtor drive or even the same GPU. It was an approximation based on what customers were generally asking about. I worked with the gear I had on hand.

In the end, what really matters is the difference from before and after changing the 3 listed components which are what people are most interested in upgrading according to our info. The PSU is not part of the equation, and not everyone needs a modular PSU. And, FYI, on the day I priced the other gear a 4GB Dominator PC8500 kit was $36 after MIR so that is certainly not outlandish. And, this can benefit people that are not even using Corsair memory. You are simply overlooking the merits of the information because you don't like something about the hardware choices.

It's already been said multiple times. This article is not meant to benefit everyone. For example, I could pick up a magazine or read a web article on how to install a Paxton supercharger into a Mustang GT. But, I don't have a Mustang and can't afford the car or Paxton. But, it is a cool upgrade and my socio economic level and vehicle needs do not diminish the validity of the upgrade article. Take it, or leave it for what it's worth.
 
Well if someone built a Phenom I system just when the CPU launched, they would have had something similar to the "before" system in the article. The 4800-series hadn't launched then and the first batch of Phenom CPUs wouldn't reach much more than 2.5 GHz much of the time anyway.

So for anyone who bought into the original "Spider" platform and is now asking if it's worth it going to the "Dragon" platform (4800-series and Phenom II), the article might be useful.

Performance wise, my system sits pretty much halfway between the two systems. I get 12k 3DMarks and a CPU score of about 3750. So the article doesn't tell me that much. I do know from using my system that it performs adequately at this point. If I were to upgrade something, I'm not even sure whether it would be the video card or the CPU - so I'll just keep both for now and replace them at the same time later.
 
Back
Top