Bike component benchmarking?

Kiblams

As anyone who is into their PC gaming will be able to tell you, computer hardware is constantly benchmarked as part of almost all reviews. Due to this is it very hard for personal bias to factor into the top hardware.

Does anyone know if it is possible to gain a similar unbiased benchamrking system to work with bike components? or if someone is actually doing this somewhere?

I am looking for new forks and am constantly having to weigh up peoples personal opinions as everyone has their own take on performance of a fork, would be great if I could see cold hard facts on wihich to base my purchase.

I know the stats will have to have a threshold as complicated engineering is never completely perfect from item to item, but even an idea of these thresholds could be good to know as I am sure some companies are more accurate than others.

Might be worth us discussing what can be measured for the benchamarks also, this may need some precision equipment to measure (in the example of forks) compression at certain PSI, rebound speeds, flex under certain forces?

These are just a few ideas I would consider. What do you guys think? Am I the only one sad enough to want these figures?

The Spiderman

MBR did a very thorough test recently of shocks in which various,factors were measured in a lab.Not sure if they have yet done a similar test on forks.

Personally,I go by reputation and my own feel of what`s right.

nicklouse

too many variables.

a fork may be great in one setting but may have some characteristics that are not great in another setting. (harder spring and heavier oil etc)

shocks... many things can effect the action of a shock like the ratio of the suspension and the arc of the acting linkage.

A lab set up testing can be so false.

Kiblams

But you also have many factors on computer benchmarking, in the instance of graphics you have resolution and game settings, and this is before you go anywhere near the driver settings. What they do is set a standard resolution for the comparison with settings mirrored accross the tests.

So surely you can chose standard oil accross forks and a standard PSI for air sprung, then choose a 'medium' coil strangth for each brand which will also be a good indication of how each brand clasify coil strengths too?

More than one of each component will need to be tested too to give a threshold and average for each product.

Surely it is the fork performance through design and implementation that is important here as pressures can be changed? therefore by creating 'standards'; maybe 3 for each variable; like 3 set pressures, 3 set coil hardnesses and 3 thickness of oil. Then you should be able to get useful data for a comparison between fork designs at least.

nicklouse

but what is the point in that when there are so many variables?

standard oil? well when fork/shock makers start using the same oil then maybe but as they do not there is no point.

then there is the question of angle.

what angle do compress a fork at? do you push straight down or through a rotation?

Kiblams

I have just realised the pointlessness of most of my last post, of course we don't make the oils the same accross the board or the coils, just the things that are adjustable by the user (damping etc.) As the oils and coils chosen to be used by the manufacturers is all part of what you are trying to test, the components should remain as they were when they left the factory.

Nick, your point about whether to test angle of compression will be a possible two peices of data gained in the testing, if both straight down or through a rotation are important then surely both should be tested?

I might be barking up the wrong tree here, but surely spending £100 plus on the opinion of some guy who has been riding much more expensive forks (and lets face it probably doesn't care about a £100-200 censored in comparison), is daft right?

UncleMonty

Kiblams wrote:

But you also have many factors on computer benchmarking, in the instance of graphics you have resolution and game settings, and this is before you go anywhere near the driver settings. What they do is set a standard resolution for the comparison with settings mirrored accross the tests.
.

Yes but in computer benchmarking the results can be accurately measured, in graphics they use frams per second, in processors it's core speed or the amount of time take to run a complex problem, in memory it's speed & latency.

With bikes & bike parts it's all subjective

UncleMonty

Kiblams wrote:

But you also have many factors on computer benchmarking, in the instance of graphics you have resolution and game settings, and this is before you go anywhere near the driver settings. What they do is set a standard resolution for the comparison with settings mirrored accross the tests.
.

Yes but in computer benchmarking the results can be accurately measured, in graphics they use frams per second, in processors it's core speed or the amount of time take to run a complex problem, in memory it's speed & latency.

With bikes & bike parts it's all subjective

RealMan

Kiblams wrote:

Might be worth us discussing what can be measured for the benchamarks also, this may need some precision equipment to measure (in the example of forks) compression at certain PSI, rebound speeds, flex under certain forces?

So if all forks had their compression at certain PSI, rebound speeds and flex under certain forces measurements on them, you would be able to tell which one was right for you?

I wouldnt.

Kiblams

RealMan wrote:

So if all forks had their compression at certain PSI, rebound speeds and flex under certain forces measurements on them, you would be able to tell which one was right for you?

I wouldnt.

The idea was that if you have a fork and you have identified a part of its behaviour that you want to improve, you can find your forks data and then find a fork with the similar behaviour only better where you want it to be.

Of course you will need a point of reference (your existing fork) to be able to make judgements based on the benchmark data.

ride_whenever

there was a very interesting but ultimately useless article a while back for disc brakes where they dyno'd them to work out power and heat dissapation etc.

Very interesting reading but at the end of the day we ride as individuals and have preferances. With computer hardware your personal preferences don't really come into it.

Fork flex and crank stiffness have been done frame stiffness would also be an interesting one. But at the end of the day it is all about wi**y waving because it comes down to personal preference.

RealMan

And what if the other stats are different, what do you do then?


Its a very nice idea, but its just not practical.

lochussie

I don't know. With computers it's obvious that more (frames, pixels, polygons etc) is better. I doubt many people could say whether they want a slightly faster or slower rebound speed for a given psi. Interesting idea though. I agree that a lot of reviews seem very subjective and are not much use.

Tank-slapper

One of the German mags tried not so long ago. There is still some subjectivity in it though:

http://www.mountainbike-magazin.de/test ... 3845.2.htm

supersonic

Tank-slapper wrote:

One of the German mags tried not so long ago. There is still some subjectivity in it though:

http://www.mountainbike-magazin.de/test ... 3845.2.htm

Good link that! I find stiffness tests interesting to read.

I see the Tora tops the performance and stiffness ;-)

supersonic

ride_whenever wrote:

there was a very interesting but ultimately useless article a while back for disc brakes where they dyno'd them to work out power and heat dissapation etc.

Very interesting reading but at the end of the day we ride as individuals and have preferances. With computer hardware your personal preferences don't really come into it.

Fork flex and crank stiffness have been done frame stiffness would also be an interesting one. But at the end of the day it is all about wi**y waving because it comes down to personal preference.

Ah, I remember that test. What was interesting is that the Deore faired better than the Hope Mini on power and heat dissipation, weight just 15g more and cost a hell of a lot less.

What do people prefer though?

Willy waving is all too true...

Good testing is a mixture of understandable objectivity with useful subjectivity.