Complaint to Kyle: CPUs & Real-world Gameplay Scaling

Q

Quicksilver

Limp Gawd
Joined
Jan 13, 2003
Messages
139
First of all Brent, Kyle, great article. Nobody else out there is doing the kind of hardware analysis you guys do.

I got a problem with your article though. My complaint is one I have with many tech websites: WHERE IS THE MIDRANGE HARDWARE!

I think it is pretty pointless to compare a 7800GTX on 3.6GHZ to a 7800GT on 3.2 GHz

That's not a representative range of hardware choices, that's all high end shit, no matter how hardcore of a gamer you are. You have the right idea with this article, but your focus is too narrowed. You tested only one setup that was even remotely midrange, and I'd even call a 2.8 p4 with a 6600GT pretty high up there.
It's all well and good seeing how well a corvette does against a porsche against a ferrari, but I wanna throw in some cameros and eclipes too.

I wanna know if it's worth upgrading the CPU in my rig with the 9800pro
I wanna see the improvement in performance i'll get replacing my 9800pro with a 7800GTX 512 on my 1.8Ghz 3000+
I wanna see the improvement in performance I'll get when I go from 2.0 GHz to 3.4 GHz on a 6600GT on my old P4
Or how about the numbers I'll see going from my AthlonXP 2.2GHz to a A64 2.2 GHz with a comparable card.

Atleast you could have tested a wider range of frequencies. How about 2.4 - 3.4 on P4 and 1.8 - 2.8 on A64. I know sombody there in the underground bunker has the OC skill to provide that range of frequencies.

How about reviews and comparisons of gear the average man (or boy) can afford. Sure, show off the GT40 in the showroom, but give us a base of comparison. Let us know how well the focuses and mustangs do too.

Your article IS helpful, and I'm sure I have no idea the amount of time and effort expended in compiling it. But, I don't come to this site to find out what is fastest, I can go anywhere and do that. I continually come here, to hard ocp, to find out what is the best value for my limited funds, and the best value is almost never the high end gear you normaly serve up.
 
You can find all that info out on the 'net but you have to do some legwork of your own. Its well worth the hassle, though.
 
I think the "enthusiast" is willing to spend more money and upgrade his hardware more often than every other year. The 6600GT is over a year old.

I agree that the minimum AMD CPU was still too expensive. The minimum Pentium 4 was a 2.8GHz Prescott that's available online for $115 (bare CPU, retail is $129) shipped on pricewatch. The minimum AMD CPU was a X2 3800+ (retail) that is around $295 on pricewatch. A 939 A64 3000+ (90nm, retail) is $129 on pricewatch and would have been better to compare at the same price point.

What real gamer can't afford a $130 CPU and a $300 video card? :p My whole desktop computer sans monitor cost under $800.
 
pxc said:
What real gamer can't afford a $130 CPU and a $300 video card? :p My whole desktop computer sans monitor cost under $800.
Click to expand...
what about people like me, i'm still running a gf4mx :p
 
My question is why wasnt a dual core of the same speed (and cache) as the single core included?

Mostly I would be interested in results with Oblivion, but this can be done another day, yes?
 
I see this all the time at almost every hardware website: They only test the latest generation. Who around here always, consistantly keeps up with the latest generation of gear?

I have a 9500 pro on an AthlonXP 2.4, a 9800 pro on a 3.2 P4, and a 6600GT on an A64 3000 and I'm contemplateing my next video card purchace. Where can I go to see these cards lined up against the lastest video cards (or even the cards from last generation)


A "real world" comparison would have:
-one CPU from this generation (A64 2.6)
-one from the last (A64 2.0) and
-one from the generation before that (AthlonXP 2.2)

and this "real world" comparison would have
-a 97/9800 class card (extremely popular in it's day)
-a 6600GT/6800GS class card,
-and a 7800GT class card.

Wouldn't a comparison of these componants serve the enthusiast better for the same amout of time and effort spent on the solely high-end comparisson we read?
 
pxc said:
What real gamer can't afford a $130 CPU and a $300 video card? :p My whole desktop computer sans monitor cost under $800.
Click to expand...

I consider myself a hardcore gamer, and a hardcore computer user and I've never spent more than $200 on a video card, and probably never will.

The CPU to Video Card performance comparison is very helpful because I use my computer for much more than gaming. A fast processor is important to me, but what I REALLY want to know is:
at what point is spending more money on a CPU pointless becuse of video card bottlenecks?

Or at what point is spending more money on a video card pointless because of CPU bottlenecks.

It would be very helpful for me to see what a P4 @ 2.4 GHz does wih a 9800, a 6600GT and a 7800GT, and then test again @ 3.4

I think that most enthusiasts would agree with me on this.
 
I see no value in reviewing anything from the 9800 era and very little value in reviewing something like a 6600GT in this capacity. The 9800 is long past it's prime and the 6600GT is just about at it's limit for anyone who wants to game above 1024x768. Honestly, reviewing hardware that is barely available at retail is a waste of time and I think there have been more reviews of the mid-range Athlon 64s and the 6600GT than any other hardware in the history of the industry.
 
I'm agreeing with the thread starter here. I think the write up was a little too limited. I mean honestly, this is freaking HARDOCP !!!. You all do a top notch job here and that is very apparent considering the respect and popularity within the community. I've been a loyal everyday reader since 2002 and while the write ups and reviews have gotten better there are still some major shortcomings. With the current expansion of the website I can see the crew getting stressed with all the added areas to report on but dont expand if your core focus is gonna suffer. HardOCP is IMHO the best at what they do and sorry if we all seem like sniveling, unappreciative little brats. Hell, consider this constructive criticism.
 
I like that HardOCP puts out articles like this one.

That said, I'd have liked to see a single-core AMD processor in the mix. Both single and dual-core Intel processors were used; why not with AMD?

I don't have a problem with the graphics cards used, though it might be nice to have a follow-up article that does the same thing with ATI video cards (I understand that this article was sponsored by BFG and they don't make ATI-based cards). The article is first and foremost about CPU-scaling, but it might be interesting to know if results differ with non-nVidia-based cards.
 
If I was a gamer buying a new card, how could you not get a 7600 GT or a 7900 GT? You always splurge a little on the card, it's the most important part. Those two cards are amazing for price/performance.
 
This won't be exactly what you are looking for but it will give you an idea of what you can find by searching the 'net for 5-10 minutes :rolleyes:

http://www.extremetech.com/article2/0,1697,1780525,00.asp

To me it seems that if one is an enthusiast one is going to do one's own research and be up to speed rather than waiting on someone else to drop all the info you ever wanted into your lap. C'mon guys - if you really are interested in seeing how different stuff compares quit griping and complaining and do some homework on your own for crying out loud.
 
BigMacAttack said:
C'mon guys - if you really are interested in seeing how different stuff compares quit griping and complaining and do some homework on your own for crying out loud.
Click to expand...
I don't think any of this is complaining --it's what is called constructive criticism.

We all appreciate the guys at HardOCP for giving us quality articles. This thread, at least from my standpoint, is meant to tell them what we might like to see, or offer suggestions on what we might be interested in. It's one thing if all we're doing is a bitch-fest here (which is not what I'm seeing); personally I think it's a matter of showing them what we'd be genuinely interested in. If there's a reason why testing in some of these ways wouldn't matter in the general scheme of things, I'm sure Kyle will post a response --he's always been good at replies.
 
We had a BFGTech 6600 GT OC in there.

We were going to include a 7300 GS but BFGTech does not produce one, we had a reference card but it has a 512MB frame buffer, not exactly representative of what is out there in retail. We also looked at the 6800 GS but BFGTech does not produce one either.

Time was a major factor in this evaluation, it took me over a month to produce what you see here. Given the time and resources there would have been more.
 
LoneWolf said:
That said, I'd have liked to see a single-core AMD processor in the mix. Both single and dual-core Intel processors were used; why not with AMD?
Click to expand...

Agreed. I was kinda annoyed not being able to see what my single core would be like compared to if I had gotten a 165 :-/
 
Well, I understand that reviewing the latest and greatest shows people what sort of power we're seeing released. It's great to know that the 7900GTX beats out the 7800GTX by "x" amount.

Now, I think the best way of doing this sort of "multiple level systems" rating, is to have the top or most popular card of that year or of a series. Such as the 9800Pro. The card was pushed with HL2 and was the top card of it's time. Then the Nvidia 6800GT series was the top overclocker closely compared to the X800Pro/XT series. Then it would be the 7800GT/GTX series. And now it's a close call between the 1900XTX and the 7900GTX. Maybe the time limit should be 3 years back. Comparing the Best processor of say 2003, 2004 and 2005 against the new 2006 models.

It would be highly possible and beneficial to those that are still working with the aged equipement.
 
They have only been working on this article for how many months? And you want them to expand it to include crap that hasn't been sold for years? Hell, they never would have gotten the article done then, it is a hell of a lot of time just to include the stuff they did.

Article is good. The last two pages is where it hits the ball out of the park.
 
I think the main focus of the article was to analyze real-world performance with hard-earned hardware. You don't want to drop tons of cash just to realize that you're not getting much out of it. The dual-core comparison, for example, showed that it doesn't give a tangible advantage over single-core CPUs (in single-threaded games). A hardcore gamer, who cares more about framerates over multitasking, might save a few hundred by knowing this.

If you're going to buy mid-range hardware or below, you're already going to expect average gameplay experiences. When you buy high-end shit, you wanna KNOW what you're paying for.
 
(cf)Eclipse said:
what about people like me, i'm still running a gf4mx :p
Click to expand...
Yeah, but you have a terrabyte of memory laying around in your "evil labratory". LOL
 
I guess all I'm saying is that if I could see the numbers the new stuff puts up compaired to my own, and I could then quantify the percentage of increase over my old stuff I could make an informed decision as to if this generation is WORTH upgrading to, or if last generation stuff is more what I'm looking for.

And I would also like to confirm that I found the article very helpful, and I would like to see more like it in the future. Perhaps with two year old hardware in it to boot!
 
You must log in or register to reply here.
Back
Top