How about not using a percentage system, or awarding a final percentage to each card, thereby removing the ability to make a direct comparison between a product review and group test? Give the top three an "award" (however you want those to be named, such as Platinum, Gold, Silver or whatever!). Each item then gets a rating breakdown for the tested areas (out of five), allowing a finer discrimination of the results between products? It might make it a bit harder for readers to pick between the fifth ranked product and sixth ranked product, but then if we're honest, people usually read a group test with a view to purchasing one of the top three anyway. I'll try and swing by my local CPC stockist this week, though my schedule is bonkers. Why oh why am I working in A&E on Friday night in a very very large teaching hospital and regional trauma centre?
Yes, I can count. My point bringing the roundups up was that those are the things I am responsible for in the mag, so for the last two related to coolers I explained the position we took on AMD. Standalone cooler reviews (and CPU too) are not something I do for the mag these days so I'm not going to talk about its position when I don't know all the facts. It's that simple. I in no way ignore AMD, and it is not my magazine. They outnumbered Nvidia hugely in the GPU labs, and I always strive to cover their latest GPUs too for standalone stuff. Again, GPUs are something I'm responsible for so I can speak on that, but I can't speak more broadly about other areas of coverage. It's not ignoring it, it's not wanting to speak from a position of ignorance, or at least not being part of the decision process. And as I said, I have informed people better placed to discuss it with you. I think that's fair, I don't know why you're having such a go at me tbh. Looking at everything from every point of view is impossible in a space-limited magazine, and you know it. I do try to cover the most relevant things for our audience in all things that I cover, however. In case that isn't clear, that is: group tests (Labs section) of all flavours, graphics cards, and also SSDs. Those are my primary responsibilities. If you think I am neglecting AMD, or other vital information in those areas, please do let me know. Outside of those, you will be best off directing your thoughts at someone else, or at the general mag feedback forum.
As I've already mentioned I'm not having a go at you but the mag. I don't know you at all and therefore wouldn't attack you. I just don't understand how a magazine dedicated to pc tech can drop something as simple as cpu cooler testing for AMD cpu's. Don't see me as bad guy I'm just asking questions. I thought I was just in a debate, my appologies.
The final percentage (in both group tests and standalones) is made up of various contributing areas, with various weightings based on their perceived importance. i.e. 1080p performance was more important than 4K performance (25 percent vs 10 percent of final score) since it will be more important to the majority of people. These weightings are explained, and the individual scores are show for each card already. They're also very different to the score breakdowns in standlones; one of the ways we try to show that direct comparisons shouldn't be made. Your job sounds far more stressful than mine; I do not envy you!
Mark just find a good 690 put it under water over clock the nuts off it an you'll be munching 980's... Sorted Well thats what I'm thinking about doing.. Always wanted a pair in SWLI just for the hell of it. Poor bloke only asked for some advice and started world war 3 on here !!
Fair enough; in that case I apologise for taking it personally. Based on my contribution to cooler testing, it was my understanding that for standlone reviews we will test on AMD FM2+ for low-end/mid-range air coolers (i.e. the most relevant products to that platform) but not for higher end air coolers or with any liquid coolers. If that has changed I did not know so can't comment on the decision - you'll have to wait for someone more informed. I will nudge them to respond if nothing is said within a day or two.
Depends on your resolution. The 2GB frame buffer of a 680 would limit a 690 at high resolutions (mainly 4K). We did 770 SLI testing (remember 770 is just a faster 680) and found its scaling dropped massively when we went to this resolution. Sadly, we did not have any 4GB variants to test with to see if this was the only limiting factor, however.
I have had a VERY busy day today, hence the clipped response earlier. Apologies for the (not deliberate) misleading summary of the 970 review. It really wasn't malicious in intent. I would actually prefer an arbitrary single score like the old Crash magazine. Please don't judge me too harshly, I'm just a washed up 46yr old bloke trying to get through the days with as little stress as is possible.
Don't sweat it. I know there was no malicious intent, just try to be careful when summarising views, as doing it inaccurately can be bad news. It's something we're all guilty of at times, but it's just all too easy for people on a public forum to read something, take it as gospel then form and share opinions based on it. Just glad I was able to respond early. Take care buddy
You aren't I'm afraid. "I'd rather read my wifes chat magazines" is not a question. If you are going to 'just ask questions' you need to alter the phrasing and tone of your writing style so that it appears less critical and aggressive. In turn you'll find people's responses are much less critical and aggressive. "I thought I was just in a debate, my apologies" - patronising (the irony is not lost on me), passive aggressive and not taking responsibility for what you have previously said. Remember, what, for you, is a simple criticism of something you've disliked is, for someone else, a criticism of hours, days and possibly weeks of hard work.
Matt's let me know about this post and thought I'd hop in on the AMD front. In short, we don't review many AMD CPUs because we don't get sent them - I'm talking specifically about the 9000-series here. AMD 'preferred not' to send us one for the initial reviews and again for the Haswell-E launch and I'm happy to say this because you're asking me directly (in fact they were only shipped to system builders and any reviews that were out there were by people that went and bought the things...). As for the 8000-series, not much has been released but we did look at the 8370E recently - again not great, and most online reviews will agree. In short, the likes of the FX-6300 and FX-8350 are still the chips to go for and they're still very much available too, if you're going for a budget system and don't mind the added heat from overclocking them - we looked at both of these a while ago. Even at the budget end, as several companies have introduced BIOSes for many non-Intel Z97 chipsets that allow you to overclock multiplier-unlocked CPUs on super-cheap boards, namely the Pentium G3258, which we recommend over something like an FX-6300. APUs we've covered most things I believe - these are more competitive against what Intel has to offer too. CPU cooler reviews are different - a while back our Phenom X6-based cooler rig died and we took the decision at the time that because we didn't recommend AMD CPUs in any of the Elite List PCs, we wouldn't bother creating a new rig and going through the hassle of testing if no one was going to be interested. Of course since then, some of the lower-end CPUs have gained a following. We initially attempted to go with APUs but they had issues with inconsistent thermal results partly due to the fact they have a hefty GPU embedded but in short the temperatures were all over the place. I believe we even looked at using an AM3 system in the last group test but unfortunately the motherboard died just as we started testing. Otherwise we would have AM3 results. Matt is our labs man but I think we should be able to get an AM3 rig sorted for the next cooler group test, after which we'll have some comparison numbers to go forwards so we can start including AMD results in future reviews
Oh wow, you're totally right, I completely forgot about this. We did indeed attempt to get AM3+ results; I took the time to set the system up, then the CPU died, and it was too late in the process to get a new one and start again. It really is something we want to do, we just haven't had the best of luck, to say the least.
Not just me then, I too believe it is pretty much the same each month these days. Considering the amount of kit out there why is it necessary to review GPU's nearly every month?
It may have read to you as patronising but it was a genuine apology as Matt will probably tell you from the long PM I sent him before you posted. I also didn't realise I was being passive aggressive, I don't have a degree in English or anything so I just type what I think and it probably came across badly.
Right up until it runs out of VRAM which sadly will become more and more common into the future. Also certain games just won't play ball with the card, either turning it into a 680 thanks to no SLI or, in the case of Titanfall, a brick due to screwed up SLI support. I adore my 690 but realistically I will need to replace it soon (well replace as in not game on it but it isn't going anywhere).
I'm think along the lines of the 970's or 980 for the warranty with EVGA now. AMD lately cards are not lasting that long so the warranty is becoming more important especially with the amount of money we spend.
Well to be honest, since the launch of the HD7970's in Jan 2012, and now with my R9 280X's (Rebranded HD7970 Ghz) - They still max out every game with ease. Not bad for a pair of GPU's that are 3yrs old. I don't understand why people worry about warranty, as most people upgrade within a yr or so. Most manufactures come with 2yrs warranty these days. As for the upgrade from your SLI GTX 770 Kovoet, I would hold on to them, as they're no faster than my pair of R9 280X's. If I can max out any game, then I'm sure you can. Right now, I would save your money, as the performance boost is minimal for the extra outlay. I'm tempted to wait till next yr when we start to see GPU dies using 16nm wafers, as any new gpu that comes out this yr will still be based on the ageing 28nm die.
