Are the review scores to high?

[DaMulta]

I always see them in the 9.0 to around 10 almost evertime. Which leads me to the point are the scores to high and or should the rating system be looked at?

[Jimmy 2004]

I always see them in the 9.0 to around 10 almost evertime. Which leads me to the point are the scores to high and or should the rating system be looked at?

I've got to agree with that actually - I've noticed most things reviewed on this site seem to get somewhere in the 8-10 area. The problem with altering the scoring system would be that the older reviews wouldn't be comparible to the newer ones because they would be on a different scale, but I definately see where you're coming from on that.

Edit: obviously it could be that the stuff being reviewed genuinely is very good but some of the scores come across as a little too generous sometimes.

[bruins004]

I have seen an ocassional 8 or 7.5, but I agree.
Everything seems to get a good score.

[W1zzard]

i dont think there is any objective way to give such scores .. read the text. use your own brain to build an opinion.

typically scores are in the 7 to 10 range. thats correct. name me a single product that would deserve a score of 3 or less in your opinion and why

[bruins004]

The Ultra X-Connects :)

[W1zzard]

and why?

[Jimmy 2004]

i dont think there is any objective way to give such scores .. read the text. use your own brain to build an opinion.

typically scores are in the 7 to 10 range. thats correct. name me a single product that would deserve a score of 3 or less in your opinion and why

In all fairness nearly all of the reviews I've read have been excellent and I agree that in the end you have to build your own opinion.

I suppose that people will just have to recognise that even though most of the reviews have quite high scores they are pretty unbiased and you can make a comparison between two reviews on the site using the scores.

[bruins004]

The 12V rail was around 50 percent efficient and the parts were very cheap inside.
Due to this a bunch of them blew up.

However, they sent some reviewers ones with high quality parts and that is why some of the reviews for these are great bc those PSUs were (the select few they sent out).

But I know you didnt review those.
But it would be neat to see a nice round up maybe.

EDIT: Keep up the good work on the reviews though. They are very thorough.

[DaMulta]

I'm not saying that I do not enjoy reading the TPU reviews. I like them a lot actually. I guess I'm just saying that I think its easy to get a fairly high score in the end.

Take this review by wiz(Not picking on you)
http://forums.techpowerup.com/showthread.php?t=17931

It wasn't the best memory for the money, but it received a fairly really high score. I would really have to look to find a product that I actually deserved a score of 3 if there is one at all.

[W1zzard]

no worries. i love constructive feedback.

being able to run 1100 mhz is gonna cost you. suggest a kit that is similar in performance and a lot cheaper. also running 3-2-2-4 at 363 mhz might make this a good 1:1 choice for medium overclocked conroes.

[magibeg]

Not to be taken the way wrong or anything but i'm sure if w1z gave out lower scores companies would be less likely to send him new products to review :P, on a side note though if you actually read through the review and bypass the score it gives a pretty accurate description of the ups and downs of the product. Rating is subjective so read the review and give it your own rating if you need one that badly.

[Namslas90]

If it doesn't blow up, catch fire, not work, and performs within advertised specs; what score are you supposed to give it a "3"?

[W1zzard]

yeah thats exactly the problem. say a motherboard that cant run more than 240 fsb, has one pcie, ac97 onboard and one 10/100 ethernet. but costs $40. as long as the board runs stock stable this might be a great choice for someone who doesn't plan on overclocking and just wants a cheap computer. i would give such a board easily a score in the 8+ range because it has potential to be useful to A LOT of users in the low-cost board market. from our pov this might be a niche market, but in reality this is the biggest market and we are in the niche. of course i would point out the limitations in the review so everybody can decided how useful the product is to himself.

i agree that the "company" argument is valid to a certain point. but i would never give a higher score or lie about a product just to make a company happy

[Namslas90]

I agree, The scores are appropriate for the reviews being conducted, no one can project a future problem that doesn't occur during review.

[DaMulta]

Not a whole lot cheaper but I would think they could hit those speeds if not better.

http://www.newegg.com/Product/Product.asp?Item=N82E16820146677

[Jimmy 2004]

I guess I'm not as interested in a score for a conclusion as much as a brief desciption of the good and bad, which this site covers quite nicely with the pros and cons bullet points. Perhaps you could go back to how the reviews used to work and give individual scores for performance and value ect? That way it would be easier to compare the performance of cards - ie. the X1300XT and X1900XT have scores which are only different by 0.2 (I think) despite the X1900 being much more powerful. This is obviously because the X1300 is good value compared to other cards, but if there was an individual score for performance people could immediately see how one compares to the other in terms of gameplay. Admittedly this is not a good substitute for reading the review altogether, but it would be useful for quick reference.

[DaMulta]

yeah thats exactly the problem. say a motherboard that cant run more than 240 fsb, has one pcie, ac97 onboard and one 10/100 ethernet. but costs $40. as long as the board runs stock stable this might be a great choice for someone who doesn't plan on overclocking and just wants a cheap computer. i would give such a board easily a score in the 8+ range because it has potential to be useful to A LOT of users in the low-cost board market. from our pov this might be a niche market, but in reality this is the biggest market and we are in the niche. of course i would point out the limitations in the review so everybody can decided how useful the product is to himself.

i agree that the "company" argument is valid to a certain point. but i would never give a higher score or lie about a product just to make a company happyYou two just made some really good points.

[W1zzard]

I guess I'm not as interested in a score for a conclusion as much as a brief desciption of the good and bad, which this site covers quite nicely with the pros and cons bullet points. Perhaps you could go back to how the reviews used to work and give individual scores for performance and value ect? That way it would be easier to compare the performance of cards - ie. the X1300XT and X1900XT have scores which are only different by 0.2 (I think) despite the X1900 being much more powerful. This is obviously because the X1300 is good value compared to other cards, but if there was an individual score for performance people could immediately see how one compares to the other in terms of gameplay. Admittedly this is not a good substitute for reading the review altogether, but it would be useful for quick reference.

we used to have such a system on my last site ocfaq. i found coming up with several scores in several categories is quite hard. also it encourages people to do number to number comparisons and not read the review, which can often be misleading.

[zekrahminator]

Here's a good reason to give a review high scores- big companies are more willing to give us "review samples" of super hardware if they know it'll get a great review :).

Of course I know that the real review score has nothing to do with how much we want more hardware. It really is a bit more like what W1zzard said earlier. As long as something has redeeming qualities, we'll be happy to give it a nice score. And the reviews that really test a product to it's limits are always fun to read. Like the USB Kart one where the reviewer (W1zzard?) ran over the USB Kart with a car until it didn't work anymore.

[Grings]

if i were to review gear (would'nt it be nice, being sent all that nice top end gear) i'd probably use a percentage instead, but reading the review itself, and the pro's and cons of hardware is far more informative than a mere score, as this is usually based on value, for example if a graphics card got 7/10 at £280, it would be a 9/10 at £200

[Darksaber]

Well as W1z mentioned, there is no real system to the scores.

The lowest score I have given is 6.2 and there was enough reason IMHO to do so.

At times items do not get a score at all, and this may be an advantage or disadvantage for that product.

Most products do meet their advertised specifications and as long as the price is justified for that performance there is no reason to give them a really bad score. If a product is completely standard (incl. cost) and performs as it is meant to, it will not get a bad score.

The items we review here are very interesting, rare, unique and useful things. In other words - stuff you guys may like and be interested in. Who wants to read about an OEM case? ;)

So you can take a score of one case and compare it to the score of another - saying "this is better than the other". This can only be done to a certain degree. If both cost the same are the same size, have the same outstanding features and are made of the same material, such a comparison may work - but how often is that the case with cases (hrhr all the pun intended).

cheers
DS

[zekrahminator]

Ah, here it is, the Corsiar Voyager review. (http://www.techpowerup.com/reviews/Corsair/FlashVoyager1GB/4) Lucky W1zzard ran his USB stick through the dishwasher.... Can I review some flash sticks? :roll:.

[Jimmy 2004]

Ah, here it is, the Corsiar Voyager review. (http://www.techpowerup.com/reviews/Corsair/FlashVoyager1GB/4) Lucky W1zzard ran his USB stick through the dishwasher.... Can I review some flash sticks? :roll:.

lol at that!

Now a USB stick like that certainly deserves a high score!

[DaMulta]

http://www.techpowerup.com/reviews/Corsair/FlashVoyager1GB/images/abuse8_small.jpg:respect:

[Easy Rhino]

scores do seem high but i think its justified. just about all the products reviewed on TPU have been reviewed on other tech sites and the ratings are all pretty congruent. i think if a product works as advertised it should get a high rating. its those that break or work incorrectly that should be given low scores. you also have to consider not pissing off the manufacturer. obviously a hairy area where fairness meets practicality. anyways, good job. i bought the ocz gamexstream 700 watt psu because of the review i read on here and it has been fantastic! :toast:

[Wile E]

Most of the reviews I've read seem to be rated fairly. I agree that as long as the product performs as advertised and is in a fair price range for it's market, that it should receive a high rating.

[Grings]

also it's a hardware enthusiasts site, why would they review crap stuff?

[Urlyin]

I can say we have had several conversations about review ratings and you also have to take into consideration that not all products review the same as for an example, trying to use a rating system that is used for a case on a video card... some of the score is based on the reviewers satisfaction with the product and I can assure you we don't make a review nice to keep the manufacture from being pissed off at us, I have seen where W1zz has pushed back and said fix it...

All in all though I can say this for myself that I enjoy doing reviews and again as W1zz has said to rate it in terms of if you would suggest your friends to buy the product or if you would spend your hard earned money on it, having said that you can get a better idea that the scores have a thought in mind and generally reflects the overall experience the reviewer had with the product as well ... we're consumers too!!