Firing Squid Wastes their Ink

p4l1ndr0m3 said:
You are completely missing the point... Here it is, cut and dry.

Firingsquad gives you 95% facts and stats and 5% opinions
Hardforums gives you 50% (of what they call facts) and 50% biased opinion.

So, if you like being told what to do because you have no brain, pick hardforums and forget about firingsquad. Otherwise, read firingsquad, read other review sites, even read hardforums, but make no one website your bible you dimwits!
Click to expand...

In an incredible twist of irony its you that actually completely miss the point. Ever have somebody ask you for computer advice because they just don't have all the facts? [H] is actually educating, not throwing a bunch of potentially nonrelevant information at you.

Joe asks: "Should I upgrade my 754 to a 939 processor?

Standard Benchmark mentality: "Yes because they use dual channel memory so its faster." (e-penis awwwwayyy!)
[H] mentality: "That depends. What do you do with your system, you may not see any performance increase whatsoever for your money."

The whole point of coming to the [H] is getting advice just like computer noobs come to an enthusiast for advice. Because the enthusiast has a wider vision, more information, more experience. Despite the fact that I deal with hardware on a daily basis doesn't mean that I can deal with it like the [H] does. I don't have access to the hardware like they do, or the games to test, etc. So do I want their opinion? Damn right I do. I don't blindly come here to be told what to get, I come here because I TRUST the advice given. Any shmuck can go read a canned benchmark review and see a 30% increase in a chart and think "Well its faster in Half-Life, so it must be faster in EVE Online."

The [H] truly explores whether or not its truly faster in as many scenarios as they can to EMPOWER the enthusiast to make a smart buying decision. They go above and beyond. If all you care about are benchmarks, just install 3DMark and dont bother visiting anymore hardware sites. I mean you already know the new stuff is gonna be faster, because hey its new. It has to be. :rolleyes:
 
Stinkfist said:
I hope and pray for the day that we become cpu bottlenecked. Graphics solutions have NEVER been up to par with cpu performance. Of course, you can bottleneck your gpu but you could never do it with current cpu tech. As it stands, most people will NEVER be at the point that they need more cpu power, that has been true for years. Seriously, how much power do you need to run a web browser, and how much difference does it make to you that you can decode, reencode and burn a dvd or mp3 file 30 seconds faster than you could before? I know time is money, but DAMN.
Click to expand...

In DVD and mp3 related stuff, even ten seconds can be the kicker for some.

To be perfectly honest though, canned or real world, what I'd like to know is how hard I can push the damn thing before overclocking it. Quad-SLI rig tests don't do dick for me. Neither do SLI ones because hey guess what? I'M STILL ON AGP LIKE A LOT OF OTHER FOLKS ARE! But at the same time I don't plan on staying on AGP much longer. I also don't plan on investing more than $1500 on my next build.

But it's still nice to see the numbers, I do admit it. However, I also like to get an idea of what I can do on the hardware if I get it. Maybe down the road I might throw a couple hundred at my rig to make it just that little bit better. Synthetic benches aren't going to tell me shit about what I want to know. And no, real world won't tell me -everything- I want to know, but y'know what? I can't also go down to Fry's and test out various components myself to see if I like their performance. Unfortunately PC parts aren't like cars on a dealership lot. No one is going to toss me the keys and let me take a $500+ video card out for a spin. That will probably be the only way anyone of us will get a true real world perspective.
 
OldPueblo said:
In an incredible twist of irony its you that actually completely miss the point. Ever have somebody ask you for computer advice because they just don't have all the facts? [H] is actually educating, not throwing a bunch of potentially nonrelevant information at you.

Joe asks: "Should I upgrade my 754 to a 939 processor?

Standard Benchmark mentality: "Yes because they use dual channel memory so its faster." (e-penis awwwwayyy!)
[H] mentality: "That depends. What do you do with your system, you may not see any performance increase whatsoever for your money."

The whole point of coming to the [H] is getting advice just like computer noobs come to an enthusiast for advice. Because the enthusiast has a wider vision, more information, more experience. Despite the fact that I deal with hardware on a daily basis doesn't mean that I can deal with it like the [H] does. I don't have access to the hardware like they do, or the games to test, etc. So do I want their opinion? Damn right I do. I don't blindly come here to be told what to get, I come here because I TRUST the advice given. Any shmuck can go read a canned benchmark review and see a 30% increase in a chart and think "Well its faster in Half-Life, so it must be faster in EVE Online."

The [H] truly explores whether or not its truly faster in as many scenarios as they can to EMPOWER the enthusiast to make a smart buying decision. They go above and beyond. If all you care about are benchmarks, just install 3DMark and dont bother visiting anymore hardware sites. I mean you already know the new stuff is gonna be faster, because hey its new. It has to be. :rolleyes:
Click to expand...

Honestly, I said this in an earlier post, if you only use info from one site to make a purchase decision, that is not wise. I personally want consistency between tests, as well as, all pertinent info about the particular piece of hardware being reviewed. I don't have a problem with reviewer's giving advice, but ignoring the fact that a CPU will show performance difference's before you get to the GPU bottleneck is retarded, period. One other thing you don't hear about much is power consumption by your CPU. Will your CPU save you money over time vs another? Probably, if you compare an Athlon64 to an Intel P4. What else does this mean? Your smaller PSU will probably be able to keep up better with the less power consuming CPU due to the smaller wattage requirement. Nothing wrong with advice, but you have to remember that enthusiasts think OUTSIDE the box.

Also, regarding your sock 754 to 939 comment. 939, because most people who have to ask about the difference aren't one to upgrade very often. So, to compensate for the potential requirements of future software, go with 939. Also, drivers will be more current as the tech isn't nearly as dead as 754's is. Also, the range of CPU's you have to choose from is much greater as well as the quality (i.e. A64 3200 754 vs A64 939).
 
p4l1ndr0m3 said:
Honestly, I said this in an earlier post, if you only use info from one site to make a purchase decision, that is not wise. I personally want consistency between tests, as well as, all pertinent info about the particular piece of hardware being reviewed. I don't have a problem with reviewer's giving advice, but ignoring the fact that a CPU will show performance difference's before you get to the GPU bottleneck is retarded, period. One other thing you don't hear about much is power consumption by your CPU. Will your CPU save you money over time vs another? Probably, if you compare an Athlon64 to an Intel P4. What else does this mean? Your smaller PSU will probably be able to keep up better with the less power consuming CPU due to the smaller wattage requirement. Nothing wrong with advice, but you have to remember that enthusiasts think OUTSIDE the box.

Also, regarding your sock 754 to 939 comment. 939, because most people who have to ask about the difference aren't one to upgrade very often. So, to compensate for the potential requirements of future software, go with 939. Also, drivers will be more current as the tech isn't nearly as dead as 754's is. Also, the range of CPU's you have to choose from is much greater as well as the quality (i.e. A64 3200 754 vs A64 939).
Click to expand...

Instead of "all that real world info" why don't I just show them a benchmark or two and let them decide? :p
 
I personally think that there's still a lot of use left in "canned benchmarks" if they're done correctly.

As far as a crystal ball, well canned benchmarks done in the right way are going to be a better indication of your hardwares performance in different circumstances, if you can take lots of benchmarks that are intensive in differen't ways, and then measure video card performance at doing those things (maybe testing fill rate, memory bandwidth or shader calculation) then when a game that comes out that heavily uses one of these features you have a better idea of what it's going to run like (assuming no benchmarks have been done for that specific game)

The problem HardOCP faces is that your articles may tell us more accurately what the gameplay experience might be like with oblivion, or half life 2, but what about the 1000's of other games out there?

The results from "canned benchmarks" can give you a rough idea of how a video card is likely to perfom if you know what sort of features the game uses a lot of and what it uses little of.

The next problem HardOCP faces is that of course they can pick and chose what "results" they show us, and since their results paint a very narrow picture only really letting us know what to expect from a few games that can throw our expectations of their hardware.

The article trys to suggest that they have an agenda and you don't, and that they make money and you don't, and that they can skew your vision about hardware, and that you don't.

Both as bad as each other IMO
 
I see the concept you are going after, however I'd put it another way. They strive to give you information that is relevant today. Right now. The other side of that coin as you say is it "can" be possible to extrapolate how well a card might run in another game using canned benchmarks, based upon whether or not its D43D, OpenGL, the Unreal engine or the Doom3 engine... like the canned benchmark that was run. However if you think about it there's almost more potential for miscalculation with the latter because you are assuming and guessing how well something will run. Engines get modified and task your machines more, drivers are buggy in some games but not others, etc. What the [H] is doing is giving you as pure a performance demonstration as anyone is able, and its info you can use immediately. You don't buy a system because you want it to perform a certain way in a year with a game that hasn't even come out yet. Look at all the suckers with Doom3 and HL2 that bought early and new hardware came out later before the games.

Its real world info you can take with you along with your real world money down to the store to buy something you know is going to be faster then what you have. TODAY.
 
RedStarSQD said:
less than 3% gamers use 1600x1200 and up ...less than 1% use amd fx 62 ...and less than 5% of gamers use 2 gig memory.

and as already stated i have no trouble with weighting [H] viewers 7x more in the high end ..still only 21%.
Click to expand...

I'm getting more and more tired of seeing that data, where ever it comes from, Steam some keep saying. Look guys. There are gamers, yes. There are 13 year old kids playing on daddies old off-lease desktop he got from work, hacking CS:S or poking at HL2. Absolutely.

But. There are gamers, and then there are [H] gamers! Googling was extremely lacking in hard data applicable to this debate, but gaming sites polls suggest in same cases 10-16% use of 16x12. These are gamers, they read about games, dream about games, drool over magazine pics of hi-rez screenies, then buy the games.. but don't know hardly jack about hardware beyond what corporate propaganda they hear, so yes, they too are no where near [H].

Forum searches turn up no previous poll regarding resolutions on Hardforum.. Though a bajillion posts suggesting that either the vast majority either play at 12x10 or 16x12 or at the very least try. Almost nobody uses 800x600, though it does seem a pretty common theme to have around 50% using 1024x768 on most typical sites. Again. [H] is ten lightyears from typical.

Whatever the case may be, comparing the mass market and thinking it's at all representative of an extremely niche computer hardware overclocking website is like saying that because most of Earth's population seems to like soccer, American's must too. Heh. No. We need a poll, seriously after this blows over to make sure its fair, but I think by and large we frag our n00bies at higher resolutions.
 
Going to have to side with FS on this one. The FS approach is more scientific and comprehensive. They strive to eliminate all outside variables so that the observed difference can be attributed to the variable that is being tested. If someone wants to believe that a 30% performance increase at 800x600 no AA/AF translates to a 30% performance increase at 1600x1200 4xAA 4xAF then let them. As far as I'm concerned, the reviewer's job is to provide empirical, quantitative evidence, with minimal interpretation. The reader should decide for himself if the part is right for him. [H] reviews nowadays make this very difficult as not all test setups are running under the same set of conditions. While it's true that running under the same conditions is not always possible, reviewers should still try make them as similar as they can. In the case of video card reviews where the same settings can look different depending on the video drivers, screenshots should accompany raw data so people can decide for themselves which one looks better.

I appreciate all the work [H] staff have done over the years, but the core 2 gaming performance review doesn't really say much. The entire review could be effectively communicated in a single phrase: "At higher graphics settings and resolutions, most games are GPU-limited." For most enthusiasts, this is common knowledge.

The wiggly line graphs also do nothing for me. It's hard to say if one line is more variable than another, or if one line is systematically higher than the other. If you want a simple, effective way of communicating the central tendency and variability of a dataset, give the average and the standard deviation. It's quicker and easier for both the reviewer and the reader.
 
WuTangClam said:
Going to have to side with FS on this one. The FS approach is more scientific and comprehensive. They strive to eliminate all outside variables so that the observed difference can be attributed to the variable that is being tested. If someone wants to believe that a 30% performance increase at 800x600 no AA/AF translates to a 30% performance increase at 1600x1200 4xAA 4xAF then let them. As far as I'm concerned, the reviewer's job is to provide empirical, quantitative evidence, with minimal interpretation. The reader should decide for himself if the part is right for him. [H] reviews nowadays make this very difficult as not all test setups are running under the same set of conditions. While it's true that running under the same conditions is not always possible, reviewers should still try make them as similar as they can. In the case of video card reviews where the same settings can look different depending on the video drivers, screenshots should accompany raw data so people can decide for themselves which one looks better.

I appreciate all the work [H] staff have done over the years, but the core 2 gaming performance review doesn't really say much. The entire review could be effectively communicated in a single phrase: "At higher graphics settings and resolutions, most games are GPU-limited." For most enthusiasts, this is common knowledge.

The wiggly line graphs also do nothing for me. It's hard to say if one line is more variable than another, or if one line is systematically higher than the other. If you want a simple, effective way of communicating the central tendency and variability of a dataset, give the average and the standard deviation. It's quicker and easier for both the reviewer and the reader.
Click to expand...

I second this.FS explanation is sound and written in a polite manner.HOCP falls flat on both.
 
Hardocp needs to publish "real world" results at different settings. Not just the best setting but at least 2 settings. Super high "best real world experience" setting and then a more conservative setting. You don’t have to use “canned benchmarks.” Just add results from more settings. You don't have to run benchmarks. Run at more settings.

This is the number one reason why Hardocp reviews have been less than awesome for awhile.

The number two reason is that the graphs are unreadable.
 
I've been reading [H]ardOCP on a daily basis for about 6 years, and this is perhaps the a fitting time to share my questions to their hardware testing methology.

Basically, 'the real world testing' method is based upon the reviewers opinion of an acceptable framerate (in a given game with X hardware). In other words, you will have to trust Kyle, Steve or whomever for having the same subjective feel of a game than you.

I'm not saying their testing method is wrong, but its way more subjective than the synthetic benchmarks [H]ardOCP compares, and distinguish, themself from.

And its flawed on a few points.

1) Future performance: Nowadays CPUs are usually limited by GPUs - what is the performance if ATI or Nvidia releases a GPU tomorrow with 2x the performance of their recent generation GPUs? Lower resolution test may give a hint about future performance. And Conroe is definately winning those (I usually support AMD, so this is not easy to say)

2) What if the next generation of games (UT2007, Quake V etc.) stresses the CPU more than the current-gen games? Then the CPU may be determining framerate / smoothness of a game. It looks like Conroe will be faster in these stressfull situations where the GPU will not help the CPU (look at some RTS results on other sites). [H]ardOCP mentions this in their review. But it should be more clear in the conclusion, as it may sway an upgrade decision for a reader.

3) A lot of people have switched to TFT screens by now; most of these have a 1280*1024 resolution limit - [H]ardOCP should test this resolution under all circumstances to give 'a real world testing' for TFT users, instead of just testing a piece of hardware to its max playable resolution in X game. In this resolution AF and AA settings may be higher. 1600*1200 (or higher) resolution performance test is actually quite useless to me (other than provide a visual quality vs. bottleneck performance analysis at that resolution).

4) [H]ardOCPs real world testing method is not fool-proof. Neither is the 'canned benchmark' approach (which is actually a wrong name for the FS method) -
As some readers has pointed out, these different approaches to hardware testing actually supplement each other quite well. I think it would be best if either side respect what the other is doing - and react in a constructive way to the inevitable criticism.
 
Hey, i'm old, I'm a lecturer and an IT nut, so indulge me here....

All examinations prove is that an individual is good at jumping through the hoops for the exam, it doesn't, per se, provide an actual measure of how a person is going to perform in the workplace.

Likewise, what benchmarks do is provide an indicator to how a particular piece of kit will perform in that benchmark alone.

I have taught many students with low exam grades that outperform 'brighter' ones, they work harder because they have to.

Going back to the hardware issue... one of my students who left this summer built an AM2 5000 rig with a 7950 DX2 graphics card, o/clocked it and scored 2000 more points in 3Dmark06 (8200) than my X2 4400 with a x1900xt (6100) yet in "Real Life" F.E.A.R gameplay my system was just a few frames behind his and yet felt 'smoother' yep, subjective I know, but that is what this debate is about.

Returning to my original tack, when you employ someone you look at their credentials but more often than not you will evaluate them yourself or take on board what others have said about them... I know I do.

So right on [H]ocp, keep up the good work and lets have subjective, qualitative reviews.. benchmarks have their place but they are simply a tool to compare one specific aspect against another - just like exams.

PS... I may be a boring old fart (having taught CS for 30 years!), but I push my systems to the limit with FEAR, video encoding, compiling etc. AND last night I started Prey - what a blast, especially the changes in gravity and the living weapons. Do yourself a favour, if you like FPS's get it and give yourself a treat. :p Yep, subjective review, but hey - personal recommendations count for a lot (see above).

Dave <listening> Focus.... Hocus Pocus...

,Sings> Yo di do, do do-di do, do do-di do... laughter
 
excelguy said:
Going back to the hardware issue... one of my students who left this summer built an AM2 5000 rig with a 7950 DX2 graphics card, o/clocked it and scored 2000 more points in 3Dmark06 (8200) than my X2 4400 with a x1900xt (6100) yet in "Real Life" F.E.A.R gameplay my system was just a few frames behind his and yet felt 'smoother' yep, subjective I know, but that is what this debate is about.
Click to expand...

I got this one

1) Your student is obviously visiting pron sites and has a virus

2) He has the MP3 player running in the background cause hes scared of the silence in F.E.A.R. + the little girl.

3) Hes probably got his Torrent runing in the background

4) Hes got an Instant Messanger on

5) Hes younger has friends and all of the above apply.

6) His vision is better and can depict changes in FPS better than you hence making him think his game running worse but actually its faster than yours. Your old your vision cant pick up FPS changes as fast as he can.

7) He can tell this is a joke :D
 
<grins>

1) Your student is obviously visiting pron sites and has a virus

Actually, he had over 400 bits of malware and other virus related problems...

For those of you on the other side of the pond I spent 4 happy years teaching CS at JMU - man I miss the Skyline Drive :(

Pink Floyd .... "careful with that axe Eugene.."

Arrrrrrrrrrrrrrrrrrrrrrgggggggggggggggggggggghhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh!!!!!!

Syd, you will be missed, rest peacefully.
 
I flicked through pages 1,2,3 and the final page, I ran out of patience, because one thing becomes apparent. Some of you really need to CALM DOWN. This is only computer hardware! I know a lot of you guys love it, as do I, but come on, it's not the be all and end all. There are many of you out there trying to provide opinions in a fair and intelligent way, and I salute you. But there are too many of you just seeing the black and white. I've heard people say these sites are "flaming" each other. That's just oversimplification, come on! Initially [H] provided an evaluation of the new intel processor, and made some remarks about other people's testing procedures. A tad on the harsh side I might say, but they have a different philosophy, and so they are explaining why they do things this way, and why they think other methods are flawed. This is not simply "flaming", they are discussing.

I applaud [H]'s new evaluation technique, they spend the time to try out the hardware themselves, on what it will be used for - games. Then they offer their opinions based on these trials. I can't agree enough with the lecturers views above. A benchmark will simply show you how well a processor can jump through a hoop. Games show you if there is an advantage in buying said processor if you want to increase your computer's performance. Some people STILL seem to be missing this simple point. There's no use in using benchmark result numbers to say if one processor is "better" than another, it's too simple. [H] have taken the time to really evaluate the hardware, and for this, they deserve praise.

Dave
 
I found your website a little over 2 years ago. Have been a fan ever since. I visit everday to checkout the articles to see what you guys have to say, because what you say makes a lot of since. I believe in what you do and how you do it and that you have me in mind when you do it. I believe no one watches out for me like you do and I appreciate that you think I am important enough to spend the money that you do to show how parts really work. Don't stop what you do, you are very important to alot of people. Keep up the good work. :D
 
The more I read about this whole situation the more I realise that the FS review gives the same results that [H] review does, and that at settings which you'd play in the CPU upgrade basically isn't worth it. The difference is the FS benchmarks return the same results, but they retun MORE information which if you are careful about interpreting allow a much greater insight into future performance and performance in games that were not tested in the review.

It's also occured to me that essentially what [H] is giving us is a very small subset of results that FS gives us, we can use the results from FS to obtain the same conclusions that [H] has come to, only we can do it better because we don't rely on [H]'s subjective view on whats an acceptable frame rate, or whats an acceptable trade off of settings like resolution and eye candy.
 
Frosteh said:
The more I read about this whole situation the more I realise that the FS review gives the same results that [H] review does, and that at settings which you'd play in the CPU upgrade basically isn't worth it. The difference is the FS benchmarks return the same results, but they retun MORE information which if you are careful about interpreting

It's also occured to me that essentially what [H] is giving us is a very small subset of results that FS gives us, we can use the results from FS to obtain the same conclusions that [H] has come to, only we can do it better because we don't rely on [H]'s subjective view on whats an acceptable frame rate, or whats an acceptable trade off of settings like resolution and eye
candy.
Click to expand...

QFT, seriously, Kyle got owned, shoulda played out this whole episode more in his head before he went on his rant.
 
Its simple......you just have to say to yourself: what am I going to do with the data and how was it obtained.

Do I play games or do I run 3D Mark all day. Reality or synthetics.
Running data or video is like synthetics.....perhaps it has a place.
Playing games is the real world application and what most of us do every day.

I dont care about synthetics. I really dont care about actual framerates......I want to know that game "X" will play smooth and at high res on "Y" combination of hardware. If I want to look at some numbers I'll go to Anantech or HotHardware.

I started reading [H] when the Doom3 hardware guide came out......there was nothing but spot on analysis in that article and it continues on today.

I prefer the straight out, no BS information I get here. No apologies are needed.
 
Firebat said:
Wow, no matter how many times it is said people are still missing the point. Lowering the res tells us how the cpu will fare when the GPU is no longer the bottleneck, a year or two down the road. We know you don't play at 800x600 :rolleyes: .
Click to expand...
I agree with this. The GPU might be the bottleneck *now*, but a couple months down the road it will not be.
This is why running the benches at low res makes sense, it shows which processor is more powerful, which means it will have more longevity.
 
What might be interesting to see is re-running the recent Core 2 evaluation (with the same CPU's but all other hardware being current) 6-12 months from now with 'future' current games. What would those number show? Would it show anything? Am I smoking crack?
 
When I read the [H]ard|OCP review of Core2Duo, I was quite happy. I use my main power system for gaming, other than gaming there isn't anything that's power hungry I do. No encoding, video editing etc. So as far as I'm concerned, I want to see if I picked up a Core2Duo right now to replace my 3800+ X2, will I get a much higher FPS at the resolutions I game at (1680x1050, 1280x1024, etc). [H]ard|OCP review gave that to me in spades.

However I understand also that a lot of people want to know exactly how much faster the CPU is at calculating compared to other CPU's, in which case I want to see some benchmarks based on calculations (ie SuperPi). This way I know if I am, or they are shopping for a new CPU that they're getting the best CPU for their money. But I can't see how they would even want a gaming benchmark for this type of data. They'd probably want SuperPi times, video encoding times, etc.

So as a gaming benchmark, I like the real-world benchmarks; but to compare CPU's quantitively I'll look elsewhere.

edit: It seems that the biggest issue [H]ard|OCP has with "canned" benchmarks isn't so much they don't have value in showing the CPU is more powerful, but the idea that this form of benchmarking is relevant when deciding on what CPU you should use when gaming. ie These canned benchmarks should be more CPU tests, not gaming benchmarks.
 
I vote for both styles.

If I am creating a new machine and I need to pick between cpu a and cpu b then canned is best as that removes other variables so only the cpu is tested - I can then see which of the cpu's is faster and by how much. I have to buy a cpu and I am looking long term so real world benches aren't so useful - I want the cpu that will last me the longest for my money, not any that isn't bottlenecked in todays games.

If on the other hand I own a machine and I am considering upgrading from cpu a to cpu b I prefer real world benches because then I can see exactly what difference the upgrade can make to my gaming experience. The difference in this scenario is I don't actually need to buy a new cpu if it isn't going to make a difference hence real world ftw.
 
The main reason I started reading SOLELY [H]ardOcp reviews is for one main thing. Reviews are done by gamers FOR gamers. And I can't agree more that other sites with their "canned benchmarks" tend to run those bench's at resolutions that I don't use - nor have a use for so for me being a [H]ardCore Game player they mean nothing to me.

Your review on the Conroe chip in a realworld benchmark of the main games of today (and not like serious sam, or quake like other sites ran their bench's on for conroe - ROFL).

I have tried to explain to my friends that alot of the review sites out there miss some major things that I look for when reading their reviews on hardware benching:

1. Full disclosure of system used. i.e. mobo/bios ver/vid card & driver ver/cpu (oc'd or not), etc
2. Is the system they are "benchmarking" setup and configured by the reviewer? (for the majority -cept [H] and maybe 1 or two others, it seems they got their rigs directly from Intel or such cause they didn't give any info pertaining to specifics of setup.
3. Run RECENT hard system hitting games! i.e. Oblivion,FEAR, etc and run them at the highest settings you can. I want truth on how new hardware runs at highest res and settings so I get the best idea if I want to spend the moola for it.
4. Numerous other little "nit picky" things.


I have asked my friends not to believe in the "hype" of any new hardware without first reading carefully each review "with a grain of salt" (so to speak). There is one friend that is adamant that the reviews he read for conroe thus far makes him think that the Conroe will own all others in gaming -which I asked him to better read into the reviews for specifics (like I mentioned above) to make sure it isn't a biased review. But it seems my suggestion fell on deaf ears.


The one thing I truely love about [H]ardOcp is that they are willing to get down to the electron to see how things bench or work in a real world scenario either it be gaming, apps, etc. And you give honest opinions to how it it works and performs so the rest of us can get real info if we want to purchase that item. To me that speaks volumes.

~Keep up the Great work guys!
 
Name Calling is unprofessional, period. Ignore the schoolyard BS and keep on going. There's plenty of net traffic for everyone.
 
Sorry due to spurious editing once i'd fininshed my post I'd missed a bit of text off, I've gone back and edited it to make a little more sense.

*edit*

This really proves good proof of the comments I've made about [H]s results really just being a subset of FS results

fs.gif


Here we see effectivly FS doing the same "real world benchmarks" as HardOCP by running the game in settings more likely to be used by the average gamer but suppling additional information along side that.

[H] accuse FS of giving more data than conclusion which they seem to automatically associate with a bad review, i however disagree, the more the readers can apply the results of the review to their own personal situation the more likely the review will benefit them.
 
ktk said:
canned benchmarks are good for:

boasting and ego ,ie my e-penis is bigger than yours

real world benchmarks are good for:

the humble computer guru who wants a peek at what he can 90% expect if he buys the product

the guru and ego crowds are split 50/50 so no point arguing with em(the ego crowd) cus they run on 98 octane ego juice and not logic.

facts are statistics .
statistics are lies.
you do the math.
Click to expand...

Statistics aren't lies. It's how someone hides them and tries to interpret them FOR YOU that is lying. See why people are mad now?
 
JethroXP said:
I agree with Kyle's philosophy. It doesn't matter to me that at 800x600 you can get 200+ FPS with the latest CPU because I don't play at 800x600. Just think how many people would be upset if the high milage car they purchased was discovered to have conducted it's MPG testing completely stipped down with things like the doors and seats removed. Of course it's going to get better milage if it weighs less, but that isn't how you are going to drive it now is it? Test it the way it's used, we accept that philosophy for many other products, why should PC components be any different?

Given that, what I would like to see in future [H] reviews is more data points. For example, the Core 2 Duo reviews only compared two Core 2 Duo CPUs against a single FX-62. That's great if you are trying to show how the top-end of Intel compares to the top-end of AMD, and naturally when comparing top-end CPUs, the margins of performance difference are more likely to be smaller. What I'd like to see is comparisons to other, more commonly owned CPUs, so that even if my own CPU isn't represented, I could pick one that I knew to be "close" in performance for my own personal comparison of what the upgrade might mean for me. From the Articles [H]ardOCP wrote I know that the Core 2 Duo is clearly faster than what I own, because I know that an FX-62 is faster than what I own, but I don't have a clear idea of how much faster, which really matters when trying to justify to the wife the expensive purchase you want to make. More data points that show how the Core 2 Duo compares to older CPUs would have made these articles perfect.
Click to expand...


I currently have an older system using a Prescott P4 at 3.6 GHz, 2GB of good low latency DDR, and an ATI X800XT AGP. It needs a graphics update (Obviously) but Kyle has shown me with his review methods that as far as gaming goes, upgrading to a better processor really won't improve my game much. Now unfortunately, to upgrade my video, I have to go PCI express which means if I want to optimize my purchases, I must choose either AMD AM2 or Core 2 Duo and get new memory as well as processor and mobo. Knowing the processor wont make much difference right now in gaming, I look at Kyle’s other Core Duo reviews to see if their is any other reasons to go Core Vs AMD. I find that the Core duo spanks AMD at a cheaper price point (until the big cuts at least) for video editing, which I also do on this system. I wouldn’t have come to the same conclusions using most of the benchmark related reviews out there. The only thing I could ask for in Kyle's reviews is, as the quoted poster said, is a comparison with other existing processors so I can get a better idea of which processor to upgrade to. I will be looking for something that performs at least as well as my current 3.6GHz processor without breaking the bank. When I built my current system, I relied heavily on HardOCP and built using the Abit IC7-G. that decision has given me an excellent system for the last 2 1/2 years that, if they still sold the latest video cards in AGP, would still meet my needs.

The bottom line is that I prefer Kyle’s methods over most other site’s because he provides the information I need to make informed purchase decisions. I really think he should have toned down the criticism of the other sites, but that is part of his style and those of us that read here regularly have come to appreciate his humor.

I used to be a typical HardOCP reader that upgraded a part every 6 months and the switch to PCIe killed that. I want to keep my upgrade costs to a minimum knowing that in 6-8 months I will be buying a faster processor and a DX10 video card anyway.
 
magoo said:
Who's the "retard".......at 800*600 the dirrerence between the FX-62 and the Core Duo is 9%..........(the cost is similar too, not to my recollection 9% given the new pricing schedules.)
At gaming resolution there is no difference at all......
I not at all impressed with 9%, thank you very much.
Click to expand...
I couldn't care less if you are impressed with the Core 2 Duo's performance. Please join the conversation and keep up. We are not discussing so much the difference in CPU's and which one is better, but among other things which method of evaluating is better.
 
I think benchmark tests are useful for comparison purposes and I think that showing games at 640x480 across several processors does show you a difference between the CPU's.

The truth of the matter is that none of that really matters. None of that equates to large gains in games at high resolution where the CPU doesn't matter as much. You'd be hard pressed to tell the difference between a A64 X2 4600+ and an FX-62 in the real world.

That isn't to say that Conroe isn't a compelling product, or that Conroe isn't worth purchasing. But anyone saying that you will see massive gains by spending a grand on a processor over an existing high end AMD setup is missleading. Especially if they base that opinion off of running a host of benchmarks at 640x480 resolution.
 
Definately siding with [H] on this one. Name one person who can afford a x6800 Core 2 that is going to be playing games at 1024x768. And while I do feel that the low res benchies hold some relevant info, declaring Core 2 as being head and shoulders above AMD FX dual core procs in gaming is flat out untrue at this time. I just think that view fails to take into account the whole picture.
 
magoo said:
Who's the "retard".......at 800*600 the dirrerence between the FX-62 and the Core Duo is 9%..........(the cost is similar too, not to my recollection 9% given the new pricing schedules.)
At gaming resolution there is no difference at all......
I'm not at all impressed with 9%, thank you very much.
Click to expand...


THERE WE GO! You now like FS method! You got to make the decision *FOR YOURSELF* instead of being TOLD what was better for you.
 
Dan_D said:
I think benchmark tests are useful for comparison purposes and I think that showing games at 640x480 across several processors does show you a difference between the CPU's.

The truth of the matter is that none of that really matters. None of that equates to large gains in games at high resolution where the CPU doesn't matter as much. You'd be hard pressed to tell the difference between a A64 X2 4600+ and an FX-62 in the real world.

That isn't to say that Conroe isn't a compelling product, or that Conroe isn't worth purchasing. But anyone saying that you will see massive gains by spending a grand on a processor over an existing high end AMD setup is missleading. Especially if they base that opinion off of running a host of benchmarks at 640x480 resolution.
Click to expand...


Dan,

I think that's what everyone's been saying really. Give us all of the data, so we can make that conclusion for ourselves. Even include the "real-world" stuff to give perspective on what gains you will see if you plop the part into your computer right now. However, I think what got people angry is being told flat out without being told why, one thing is better than another.
 
Mike160 said:
Gee.. Then why don't we all wait "a year or two" and upgrade then ?? I could prob save about $300 !!!
Click to expand...

because some of us are stuck with 2.6cels and want a new cpu to last.
 
Since opinions are like a**holes everyone has them, sohere is mine. I can’t think of any site that I trust more then Hardocp, the hard work, effort and time they put into there site lets me know that if I bought a comparable system to there review I would see these numbers when I game and in fact I have several times, the Hard has never let me down. I can simply just click on anyone of their articles and read for FREE and walk away with information that I share and base my buying decisions on. Not only this but Hardocp also links other hardware sites that I could also read about what they have to say I don’t know of any other hardware review sites that do this. Because of this and many other reasons it’s my belief that the hardworking people behind Hardocp do care and want us to walk away with a true understanding of what Hardware A will do with Game B and we should expect as a true gameplay experience. Another thing I would like to bring up is that 10 years ago we didn’t see the CHEATING on timedemos like we all have seen over the past few years. 10 years ago FPS was the end all be all as long as Game X ran at FPS X then that card was the king no one really was interested in image quality. Today we have higher resolutions and larger screens thanks to newer technologies in CRTs and LCDs, image quality is just as or even more important then the old king known as FPS. Because 10 years ago FPS were so important the GREEN team and RED team would tweak, test and even cheat on timedemos, not making a game run better but making sure that timedemos would run better, the reason why they did this is because they knew testing up until “real world gaming” was done using just these timedemos so if I could make my card run 10FPS faster in Game Xs timedemo then I will be king and everyone will buy me. I hope my opinions make some sense of it all, keep up the good work, that goes out to every hardware site not just Hardocp.
 
You must log in or register to reply here.
Back
Top