Testing knives: objective results vs impressions

Alarion:

The problem with extreme testing is scale and range. Taking your corrosion resistance test as the scenario, there are steels that are highly corrosion resistant, those that are pretty corrosion resistant, and those that rust aggressively. A test such as you suggest will give data in hours that it takes a certain blade to rust, but the only really useful information to a user is whether his blade can be neglected completely, a little bit, or not at all. Seat of the pants evaluation can tell you that. In fact, a steel alloy is a steel alloy, and you can get a fair idea about corrosion resistance just based on the steel's composition and reputation and not have to do any testing at all. One knife made out of ATS-34 will perform, very roughly, like another. That will hold true except in cases where there are wide variations in heat treatment, and blade finish, which only goes further to nullify "objective" testing because those variable factors are difficult to standardize or control on actual knives.

The same is true of strength or durability testing. Comparing two blades of the same steel is pretty uninteresting. There are significant differences in edge durability between knives, but generally, a good steel, well heat treated, provides a good cutting tool. There are differences between heat treatments, primarily between factory and hand made blades, but again, seat of the pants evaluation will tell you if a blade edge is holding up to par or not. Testing two D-2 blades and quantifying the edge durability on blade A as 98.754 and the durability on blade B is 98.391 is not important. All you need is a seat of the pants test to be sure that the edge durability is in the high 90's somewhere, where you expect a D-2 edge to be.

I believe that the ultimate differences in performance between premium knives is relatively small. You can quickly learn what the limits of a good knife are, and quickly determine if a given knife is or is not up to snuff. This is why design is so much more important to me than steel type. As long as a maker is using a premium steel that fits the application, and has done enough of his own evaluating to know he is doing a good job of heat treating, then the grind and shape of the knife are much greater factors in the "performance" than anything else, and those are almost impossible to quantify.
 
Exactly because of heat treatment and surface finish I think some testing of the rust resistance of the knife, not based solely on steel characteristics is extremely important.
Take two pieces of, say, 1050.
Grind one of them with a bastard file, then take the other and polish it to a mirror finish.
Subject them to the salt water test and see how much does it take to the rough ground steel to rust, and how much more it will rust if compared to the polished one.
Even faster: take the same two pieces and, after degreasing them with acetone, dab them with cold blueing solution.
Blueing is just a form of oxidizing. You'll see the rough piece of steel getting instantly black as a coal mine in a moonless night, while you'll have to make many passes to obtain a pale bluish gray on the mirror like steel.
Even more difference there would be if one of the two pieces of steel was heat treated.

The same is true about heat treating.
As you say, same steel equally WELL heat treated will lead to similar results.
We all know mishaps happen, and anyway stating clearly which is the wear resistance of a given blade, or even testing different grinding methods will be of great help.

Yes, seat of the pants works VERY well. Take cars, again.
Sit on it. look at it. Open and close the doors, turn the eingine on and make just a few hundred meters. You will have a pretty good idea of how the car performs and how good it is.

Then: why in the world car magazines sell thousands of copies every day? :)
Because not everybody has the time to test every car.
Because not everybody can test a given car against his competitors.
Because, sure as hell, nobody will buy a car to subject it to a crash test.

I'm not speaking of custom knives only. I think such tests for them aren't even needed. If I buy a damn Rolls Royce it should damn well perform flawlessly :)
But there are LOTS of everyday knives about which we know very little, which almost never get objectively compared against their competitors and that people buy essentially out of impressions got from other people.
This may be good for the expert, or may be good in some other instances, but is a very poor service knife experts do to non-experts, especially if we speak of knife experts which write on knife magazines.

An example?
I've heard GREAT things about chisel grind, and I've heard UGLY things about chisel grind.
Now: how I can tell if chisel grind actually performs better in certain situations or it's just hype and a way to sell knives built with a far less expensive sharpening process than standard grind?
Easy: I should just buy such a knife and test it's cutting ability against a traditionally sharpened knife of same quality.
Easy?
How I can find "same quality" to start with, since there is no knowledge database regarding FACTS?
Why in the world should I risk buying a crappy knife because I can't have FACTS about a certain design, but only OPINIONS?

I'm not saying opinions are irrelevant: they are extremely useful indeed, and one couldn't and shouldn't do without them, and should always seek evaluation from somebody who did also seat of the pants evaluation of the knife. Also, but not only. One also needs facts. Just that.

I've never bought a chisel ground knife because IMHO it sucks. But it's just that: My Humble Opinion, one I formed reading and listening to many people speaking of chisel ground edges. I have no hard facts to say "No, it doesn't work as well as traditional grind". I've lost lots of time, and I have no proof, just an evaluation that may very well be wrong and precluding me from using fine knives just because I can't afford to throw away money just to give it a try.
 
Just have a look at the Balde Discussion Forums, the thread named "what's so great about Sebenza".
Here's a little quote from one of the answers, from 92degree
---
Lots of folk will tell you the Sebbie isn't worth the price; lots of folks will tell you it's a bargain.
---

See what I mean?;)

This pretty well sums it up all about the need for objective data in addiction to personal evaluation.
 
Originally posted by Alarion
...
Take two pieces of, say, 1050.
Grind one of them with a bastard file, then take the other and polish it to a mirror finish.
Subject them to the salt water test and see how much does it take to the rough ground steel to rust, and how much more it will rust if compared to the polished one...

Perfect example of meaningless testing. Its 1050. The rough piece would rust in five minutes, the mirror piece would rust in eight minutes, but everyone already knows that 1050 rusts like a mother. What more do you need to know?
 
"rusts like a mother" ?
What that means? :)
Anyway, you may not know what a knife is made of, or you may not know the efficience of a certain protective agent, or you may want to do the test the other way round, to evaluate the finish degree of a blade, or you may want to evaluate how much rust resistant is a certain knife made of a know material so to show it to people who are NOT experts and wouldn't know the difference between 1050 and 440A even if it jumped up and bite their bum :)

What you say is the same as saying: everybody knows that a Ferrari is very fast, what's the point in measuring its top speed?
Or saying: everybody knows Bang & Olufsen stereo sets are good, what's the point in measuring distortion on it?
Because it's interesting knowing it, because continual measurements show immediately lapses in production quality, or a bad steel batch (which, yes, may happen)
In the end, because it's FACT, and FACTS shut up a lot of controversy and help everyone who's not an expert to make some evaluations based on FACTS rather than OPINIONS.
 
"rusts like a mother" ?
What that means? :)

Sorry. That's a US colloquialism. Essentially, the phrase taken as a whole means rusts very quickly.

What you say is the same as saying: everybody knows that a Ferrari is very fast, what's the point in measuring its top speed?

Well, you make a valid point, but knives are much simpler objects than Ferraris. The top speed of a motor vehicle may be interesting to the neophyte, but once you know the top speed of a particular model, you really aren't interested in seeing it compared to the top speeds of other vehicles, the top speeds of which you also know, over and over. Every year Ferrari comes out with new models or makes changes to existing models, but new knife steels don't come out that often. It is interesting to see objective testing of new alloys when they do come out, but that just doesn't happen often enough to fill a monthly magazine.

In the end, because it's FACT, and FACTS shut up a lot of controversy and help everyone who's not an expert to make some evaluations based on FACTS rather than OPINIONS.

Not really. You still need the opinions of like minded people to go along with objective testing, because whether or not you like a knife depends on a lot of subjective issues, even more so than the objective issues.
 
everybody knows Bang & Olufsen stereo sets are good, what's the point in measuring distortion on it?

Never heard of them.

BTW my mother is not rusty, maybe a little dusty but certainly not rusty.

If you want truly objective tests look at those conducted at Spyderco. Ask Sal Glesser about the costs of those tests and see if that is something you're interested in doing.


Stay sharp,
Chad
 
Originally posted by Steve Harvey


Not really. You still need the opinions of like minded people to go along with objective testing, because whether or not you like a knife depends on a lot of subjective issues, even more so than the objective issues.

Yeah, obviously, as I stated in many of my posts, objective data is something you get over subjective data.
As for cars :)
You get the numerical data, you get the drive impressions and then compare this with some other typical car of that type and make your own opinion.

Chad: Bang & Olufsen are swedish made stereo sets famous for their topmost quality and design. They are also bloody expensive :(
otherwise I would have already got one! :)
Have a look at their site
http://www.bang-olufsen.com/
 
Back
Top