Obviously. But one can significantly lessen the uncertainty by having the correct specs to run the game and when you’re dealing with the kind of situation I am, any more info is good info.
-Kitsune
You could staff an entire cornfield with the strawmen thrown out by the magazine guys here.
“No way we could ever say anything whatsoever about sys reqs because every single reviewer would need 527 different machines! And you can’t give a 100% guarantee to every single reader that the game will work so it’s completely pointless anyway!”
Hey, why do you even publish reviews since you can’t guarantee that every single reader will agree with you? Absolutely pointless, right?
Look guys, all you need is a range of 5-10 machines in your office. Then the reviewer keeps a couple of saved games and checks them out on the various machines once he’s finished reviewing. Another staff member could do that, too, if the reviewer isn’t present. Then you make a little chart that shows what options you need to make the game run well, if at all possible, on a given configuration.
Believe it or not, but readers are actually smart enough to notice that you didn’t test THEIR configuration. Hardware performance for a given machine can usually be interpolated from components with a known lower or higher performance. It’s not black magic, and it’s not rocket science. Actually it’s a standard feature in German gaming magazines, as I believe has been repeatedly mentioned here and elsewhere.
Chris, a lot of reviews are written by freelancers. Freelancers can’t afford the kind of hardware expense it would take to test games on a multitude of systems.
That said, I don’t see why CGM can’t list the publisher’s hardware specs. Seems a bit silly not too, even if the publisher’s specs aren’t always accurate. There have been plenty of times when the recommended specs exceed my specs and I appreciate knowing that. The argument that the specs are on the box doesn’t help me when I’m reading a review.
If the mags really did want games tested on a spectrum of machines, the could probably pay a hardware site to do it.
I write for 'em, so disqualify this comment if you want, but I like the PC Gamer system – list the key requirements from the publisher (processor, ram, video card, HD, and that’s it – it’s kinda goofy to list CD-ROM speed, soundcard, etc.) – but then they also have a “recommended”, which isn’t the publisher’s recommendation, but what the reviewer thinks are realistic requirements for a good experience. Sure it’s a guestimate, but I think it’s pretty easy to tell if the minimum requirements are radically understated, and to give a useful ballpark range.
And I write for CG, so take this for what it’s worth, too, but I love the decision to get rid of the system requirements. Mostly because they’re meaningless and redundant. When you get right down to it, you’ve got two classes of gamer–people who don’t have a clue about what they’ve got in their boxes and will ask the guy behind the counter if the game will run on their machine, and hardcore people who know everything about what they’re running. So listing the requirements in a print review doesn’t really serve anyone. It’d make more sense for the reviewer to note this in the body of the review, if warranted. Meaning, if the game demands a serious rig, mention it. If the game’s really forgiving, mention that. You can do that briefly and it underscores the tech requirements in a more effective way than just throwing numbers in a box.
I also hate the requirements stuff because it reinforces the opinion that PC gaming is supremely geeky. That you have to know exactly what’s in your machine in order to play PC games. Which really isn’t true at all these days, as long as you’ve got a modern machine. Also, it just seems stupid. To me, it’s like listing the number of pages, weight of the paper stock used, and font type in a book review.
Sure, but this whole argument is specious. Nobody expects the reviewers to do the test series at home – judging the performance doesn’t require playing the game all the way through, after all. Nor does it require any superior judgment regarding quality of gameplay.
Like I said, provide a couple of saved games at “interesting” locations (perhaps the same ones you’d use for screenshots anyway). Then load the games and fiddle with the game settings on each test system for a couple of minutes. The whole test run should be done in an hour per game, and any editorial staffer could do it and note the results – no superior writing skills required. Grade the results on a subjective scale: unplayable, unresponsive, smooth with options turned down, smooth with all options maxed. Then put the results in a wee little table, and maybe put the box requirements below to shame the publisher.
That said, I don’t see why CGM can’t list the publisher’s hardware specs. Seems a bit silly not too, even if the publisher’s specs aren’t always accurate. There have been plenty of times when the recommended specs exceed my specs and I appreciate knowing that. The argument that the specs are on the box doesn’t help me when I’m reading a review.
I agree that listing box reqs is helpful in a print magazine since you can’t just hyperlink to the publisher’s website, but I also agree with the critics who say that these requirements have little relationship with what you really need to run the game. That’s why I suggest a test series.
If the mags really did want games tested on a spectrum of machines, the could probably pay a hardware site to do it.
Sure, that would be another possibility. Like I said, no particular game reviewing competence is required to cross-check settings and system configurations, so it should be simple enough to farm out. And for the sake of consistent ratings it would be better anyway to have the same team perform the tests for all games.
The refusal to do such tests is completely mystifying to me because this is such an obvious opportunity for magazines with a real budget and office space to set themselves apart from the myriad of free websites. Nobody notices the better writing unless they’re already reading the magazine but immediately useful information stands out, even when you just browse it.
But who the hell would care? Does anyone really want to see a magazine or website doing this? Why do you think such a table would be useful? It seems like a very tedious, time-consuming job, too. You toss it off like it’d be no big deal, but considering that most mags review a dozen or more games each issue, it’d occupy a number of machines and an employee for at least a couple of days every month. More if you’re talking about running full benchmarks.
And to what end? A table that shows green lights for some rigs and red lights for others? You wouldn’t have any room for descriptions of how the games performed, so you’d have to come up with some kind of icon system. Which would look goofy and be confusing unless you went with a good/bad deal like the green light/red light, and that wouldn’t tell the reader anything. It would also be subjective (worthless), unless, again, you wanted to make a huge table listing actual specs like framerates at different resolutions, etc. What you’re talking about seems more appropriate to video card testing articles.
Well yeah, it’s called “work”… but as I repeatedly said you’re not supposed to do it anyway. Sure, it would occupy an employee for a day or two per month. Considering that magazines already have a number of full-time employees I fail to see why that’s such a big deal.
You wouldn’t have any room for descriptions of how the games performed, so you’d have to come up with some kind of icon system. Which would look goofy and be confusing unless you went with a good/bad deal like the green light/red light, and that wouldn’t tell the reader anything.
Why not? Five systems, from 500 MHz to 3 GHz, with appropriate graphics cards (same age/price range) and RAM. The systems are described in detail at one place in the magazine. Yes, you’d have a couple of icons. I fail to see why this is any more “goofy and confusing” than those goofy and confusing stars that you guys normally attach to reviews. In fact you could use a similar system, numbers or stars to indicate performance levels.
It would also be subjective (worthless)
I just don’t get this. You mean, unlike the text of your review which is absolutely verifiably objective? If you claim that your review has any value to the reader then certainly such a performance rating would have some value as well. It’s not indended for benchmark freaks but for people who want to gauge, at a glance, how the game will fare on their system.
I think that system is fine, and it’s all I would realistically expect. Even just listing the reviewer’s machine specs in order to provide a rough frame of reference would be enough for me. Just … something…anything. Some acknowledgement that PC games aren’t console games. Regardless of the risk of looking like “geeks”, hardware is something that most people have to consider at least a small bit when playing PC games. PC game magazines should acknowledge that and make it easier, not pretend that the issue doesn’t exist.
I consider myself to be in the 2nd category, but I would still appreciate some consideration of hardware in the review. I cobble my PC together with bits and pieces over time in an attempt to keep up. In fact, I think it would serve me even more than the “clueless” types, since I could quickly understand what my chances of running the game would be. It’s easier than me having to look the game specs up on a website separately or trudge down to the store to look at the box.
I dunno guys, I review these things and I don’t even look at the box to see if it’ll run on my machine before I buy something. All I read about is whether the game is enjoyable or not if I even do that before purchasing. I think you’re all talking about wasting a lot of space in print that we could better use to tell you the important stuff…like whether or not you should even think about buying the game in question through evaluation.
Brett’s right too that the requirements imply geekiness to even bother trying to play the game in the first place on a PC. I think the CGM reviewers do a fine job of letting you know if they think system requirements are overly high in the text of the review where they matter most. We can tell you that the game runs choppily or whatever and explain how that affects your gameplay experience rather than just plopping some arbitrary guesstimate of a machine into a “recommended” box that takes up space better used for true evaluation.
Anyway, I doubt most readers care. I’ll bet Bauman can tell us that most people still just write in to ask for more previews. How many of you buy all the PC game mags anyway?
–Dave
I agree pretty much completely. I don’t have a super high end system at home, but it’s enough to play pretty much anything that’s out, so I don’t really look at the system requirements when I’m buying a game. I do, however, pay a lot of attention to reviews. If something is called out in a review as being a system hog, that’s when I’m on alert.
I’d prefer it if reviews only mentioned “recommended” specs if the game needed a powerful/unusual PC to run it well. Otherwise, ditch it and tell me more about the game itself…
No matter how careful you are or how much testing you do, you’re still going to have people who find that Railroad Tycoon 3 crashes on their system, that GTA3 doesn’t work on their system, etc., I had severe performance problems for 13 months because of a motherboard upgrade I didn’t know I needed.
I’ve reviewed for PCGamer and I’ve reviewed for CGM (before and after they ditched the sys reqs). I don’t see much value to consumers in what Desslock calls a “guestimate,” unless the reviewer making that guess is particularly hardware savvy. And by that I mean “hardware editor level savvy.” A magazine is probably much better off parroting the company line, because then at least, any error is on the game publisher’s fault. And there’s little reason to print that because it’s on the box. Maybe Steve’s right. Maybe the magazine should be solely concerned with whether the game is good and why.
I go through cycles on system requirements, mostly because I only upgrade every three or four years. So when my rig is new I look at the system requirements to see how leet my rig is compared to the game. Then on the other end I look at system requirements to see if I can still play the game. Right now I’m running a PIII 500 so I’m on the tail end of my cycle. Knowing how quickly min specs are moving also used to give me an idea of where I stood on the upgrade curve.
Either way, I used to look at the system requirements.
I read CGM every month. I miss the system requirements. Just the manufacturer’s “guesstimate”, I don’t need a freakin’ lab running tests on save games.
One thing that hasn’t been brought up is that other content will have to be cut. Magazine size is determined by ad pages, so every little blurb on system requirements is going to take away from something else on the magazine.
Hasn’t been brought up? I thought I said that in two different posts above?!
I think you’re all talking about wasting a lot of space in print that we could better use to tell you the important stuff…like whether or not you should even think about buying the game in question through evaluation.
–Dave
Hasn’t been brought up? I thought I said that in two different posts above?!
Oops, must have skimmed that part :)
Well, that’s bizarre. I’ve been burned enough that I always look, and at this point am holding off on buying most new games until I upgrade.
I stopped reading magazines a while ago, but this is the kind of thing that might have enough value that I would buy them again. (Not that reviews aren’t nice, but I can get a good sense of the game just by browsing message boards or Usenet. Where that doesn’t work is for aspects that take work to evalute.)
I didn’t realize we were underworked.
We have 25 reviews in our next issue. Installing that many games on five different systems, loading multiple saved games, playing enough to get a feel for performance… yikes.
I just don’t get this. You mean, unlike the text of your review which is absolutely verifiably objective?
The big difference is performance is verifiable and quanifiable. We can disagree on art direction, but we can agree that a game that gets 10FPS is not cool.
(Of course that’s not true for everyone; tolerance for performance varies wildly between people.)
Also, how do we test all of these games? Use FRAPS? Eyeball it? How do you take into consideration the type of game? If Railroad Tycoon 3 doesn’t run that fast on a P500, is that as big a deal as if Max Payne 2 has similar performance issues?
Or do you have to have different standards, in which you need another page to discuss why those two games aren’t handled identically.
If you claim that your review has any value to the reader then certainly such a performance rating would have some value as well. It’s not indended for benchmark freaks but for people who want to gauge, at a glance, how the game will fare on their system.
I don’t doubt it has value, at least in theory, but it’s technically infeasible, it would take way too much time, and despite all of this work, it will still be unable to answer the question of “How will this run on my machine?” Since a huge percentage of games have a demo, this isn’t even an issue for those really concerned about this kind of thing.
It’ll also create a situation where readers get even more pissed off at the publication. We already get blamed for delays, for bugs, and various other things. Not sure if I want to be blamed for a game running slower than we said it would.
I haven’t looked at Sys Requirements for a long long time - Even when I was using a P3 500 when the P4 1.5Ghz’s were out. You can almost always get a game to run on your system - you just have to dial things down to do so.
I think system requirements are meaningless these days. If your PC is more than 2 years old now then you are out of the ballgame when it comes to being able to play the latest 3D games.