A standard performance/price ratio

What we (the PC gaming community) need is a price-to-performance ratio. How much does each dollar you spend translate into performance? No BS. Just straight statistics.

We need an unbiased and accurate program (a company who values a good name more than bribes) to provide the program. 3DMark is the closest we have to a standardized measurement- however it is far from sufficient. The consumers and companies need to agree on the accuracy of the measurement (many people on this forum don't trust synthetic benchmarks to accurately convey performance).

Once gamers agree upon an unbiased system to determine a price-to-performance ratio can can we consumers stop being screwed by the baseless propaganda of hardware manufacturers' never-ending hype.

 

It's just not that simple!!

Benchmarks != real games for a start. I don't know about you but I bought my laptop to play games not for e-peen on benchmarks. 3dmark06 is especially long in the tooth, and since then they make you f*kn pay for vantage and 11 runs after the 1st so they can harvest your hardware details and sell the usage stats onto developers.

Then there's the constant errors seeping into benchmark lists like 3dmark06 runs of 1280x720 on 1366x768 panels which give a much better score than the standard run at 1280x1024 which everyone else on earth uses.

Then there's how different GPU architectures handle benchmarks vs real games, and CPU choices which can affect GPU scores (e.g. GTX260M has average 3dm06 score of 10,003 but this ranges from i7 quad-equipped M15X to underclocked GTX260M with 2.0GHz Core 2 Duo in G51VX, for example.

Notebookcheck come close with their benchmarklists, but they are of course subject to the problems I've outlined above. Plus as a reviewer they are hardly independent.

There's no substitute to doing your own research, there just isn't anyone you can rely on to do it for you. And with the nature of the internet being as it is, there would be so many criticisms and complaints you'd not trust it anyway.

 

Right now Toshiba is giving us the best price to performance ratio with their gaming laptops. No diggity, no doubt...

 

What are you talking about? There's a free basic version with most features disabled and a paid full version.

 

Vantage (at least in April, when I used it) only allows 1 use and afterwards you've to pay, it is not 3D Mark 06. I'm not sure on 11 (3D Mark 11?).

On topic: Like already mentioned, notebookcheck is the best (though not optimal by any strech) source of performance for laptops. In regards to price... it depends where you live.

A nice performance test would be: there should be 2 or 3 games that work as a standard for each direct X version (say nowadays direct X 10 and 11). These games should be tested with the same settings (obviously) and same resolution (say 1280*720 or 1280*800). And there you go for some level of 'performance test'. I know there are many variables that can be criticized (different CPUs, different video cards, amount of RAM, etc.) but I think overall is a step forward to a more standarized test for gaming performance.

What do you think guys?

 

lololololnot

 

its up the consumer to be discriminating in this respect. performance is a constantly evolving metric. look up game benchmarks, reviews, and product cost. it should be easy enough to find what you are looking for.

 

There is no way in the world you are going to convince Apple to start selling their computers for $300.

Never gunna happen.

 

In the high end gaming stuff Asus has the best price vs performance I think.

But like previously said its just not that simple to do.

So many things contribule to the price of the machine that may not contribute to the performance for starters.

Screen Quality, Warranty, HDD size/speed so right away you cant lock all those apsects out with a price ratio.

Also no matter what things always effect performance like drivers, benchmarks used etc.

If you purchased a system purely out of price vs performance the winning machine would probably be a $200 netbook

 

Read the short intro of the article.

Cars With the Most Horsepower Per Dollar- Yahoo! Autos Article Page

I think a similar stat would benefit PC gamers shopping for a prefabricated PC/laptop. It would be most beneficial in determining the value of a laptop sold without the option to customize at time of purchase. (Eg. HP, Asus, Gateway.. but not Alienware, Clevo, Sager, etc).

However there also seems to be common setups among designer laptops, so a stat for a common setup (Eg. i7 720, 5870 CFX, DDR3 4 gigs) would be of benefit to consumers.

 

Hexus does something similar to Desktop GPUs but as other people have said there are a lot of different factors that go into a good laptop. Personally I think the best way to protect yourself against hype is to do your research and make your own judgements. My computers are more than just a benchmark number!

 

History shows us problems with relying on only a few metrics...

from 2003 re the infamous Geforce FX: Further NVIDIA optimizations for 3DMark03? - The Tech Report - Page 1
from 2009 re Intel GMA: Intel graphics drivers employ questionable 3DMark Vantage optimizations - The Tech Report - Page 1

The less metrics that are relied upon, the more the reason for a manufacturer to engage in this kind of chicanery. I don't want a product that's 20% faster in a specific benchmark, I want one that's 10% better across ALL games.

Just like cars, there are many factors that a (smart) buyer takes into account when making a purchasing decision. And what each person rates on their 'hierarchy of wants and needs' is different from everyone else.


and I know it wasn't your point and it's off topic but it's exactly the same issue with reducing a car to a hp rating alone. There are other factors even just relating to engine performance/driveability, e.g.
* torque ... also kind of important in determining acceleration
* where in the rev range the max hp occurs ... (outside of a racing track) I'd much rather a broad spread over the rev range rather than an engine that's a demon at 6krpm but a slug below 3krpm (especially in a manual!!)
* how long it lasts... KIA? The most appropriately named car make ever.
* what it's like to drive, safety etc. KIA the people inside as well.
* moddability - someone interested in pure HP will most likely modify. For someone who's willing, a 150hp engine that can have a +100hp blower slapped on it is better than a 200hp that can't

 

I completely agree. You can't just line everything up and start measuring as if there is only one standard to be checked against. Something is good in one thing, another in another. We need to measure against a whole horde of standards to provide relevant results for one's needs...