Spyderco Forums

JNewell wrote:Those results don't seem to make much sense.

They are exactly what you should expect if you rank knives by a measurement which is highly variable, that variability should be reflected in the rankings. It looks like real data. If that came out in some kind of highly ordered list, correlated to some particular attribute it would just mean that the data was made up.

Given the large variations that can happen in that kind of work you would want to have it repeated about 10 times at a minimum, pool that data, do some basic normative statistics to even begin to see a pattern/correlation.

Yes, well...the data wouldn't be the only thing that accumulated in a pool... :rolleyes: :D

JNewell wrote:Yes, well...the data wouldn't be the only thing that accumulated in a pool... :rolleyes: :D

You're not kidding man that's 70 deer they processed :eek:

Yep. This is not a F&W matter, this is an EPA Superfund site. :eek:

This is a pretty weird thread you guys have going here...

I posess the M4 and I am skinz all your deer!!!

Evil D wrote: Any suggestions on how many squirrel bodies is a sufficient amount to slice through to qualify as a sharp edge?

Grey squirrels or Fox squirrels?

154CM is still my favorite.

B-N wrote:Grey squirrels or Fox squirrels?

I don't know but this is a pretty squirrelly thread. :rolleyes: Were those baby deer, mama deer, papa deer, or john deere? Oh my dear, deer. :o

Gijoe, folks are just havin fun don't take this personal. You've got to admit, some of the posts here are pretty good. I would suspect the john doe results might be somewhat accurate if the controls were similar.

Well, legend says that when a samurai sword was forged and sharpened back in the day, it was tested by slicing through a pile of bodies from your fallen enemies, and if it didn't slice clean through a certain number of bodies, the sword was melted down and destroyed and made new again. Maybe they're onto something with this skinning to test for sharpness thing. I need to start trapping squirrels or something in my back hard and then I'll test my edges by slicing through the bodies of fallen enemy squirrels. Any suggestions on how many squirrel bodies is a sufficient amount to slice through to qualify as a sharp edge?

I read they had professional testers and would give each sword a grade, like "two bodies", or "1 body", up to 4 bodies. They used criminals, supposedly dead but there is some question about this. :)

Not scientific at all, but I have found my Gayle Bradley (M4) significantly outperforms my endura4 in VG-10 at work duties. These include cutting rope, boxes, pipe insulation (rubber and paper wrapped fiberglass), hose, and other stuff I'm forgetting right now. VG-10 is no slouch for EDC use, but my limited selection of steels ranks M4, S30v, then VG-10. Actually, the S30V is no slouch at work either. I find VG-10 easier to get super sharp, but M4 and S30V do seem to stay work sharp much longer. Your mileage may vary.

S90v > m4

I got what I hoped with this thread. You know people like us who appreciate a good knife, no matter what the steel are in a minority of the buying public. People who read the magazine with small knowledge of what a good knife can do will probably go looking for a knife that has M4 and not buy it because of the price and then complain about their inexpensive tool. Observation comes more with age I believe. A lot of people believe what you tell them with no knowledge or experience to back it up. I still like my H1 Pacific Salt, serrated for my uses better than any thing so far. Good knife hunting guys and gals!
JOE

I'd like to read the article and will most likely. Prior to that though the list is endless (based on what I've read in this thread) on how the testing is incomplete. That's by my standards and my standards are low already. :) But different blades of different steels being used and counting the deer before a touch up is needed is a good enough test for me. If M4 will dress 5 deer and another steel only dresses one deer I'd say M4 is the way to go no matter what the handle is made of or looks like or even how comfortable it is. But if M4 will dress 5 deer and another steel will dress 4 deer I'd say the blade steel and edge retention is close enough (for me) that other factors would play a bigger part in the determination of which knife to buy or which one to take into the woods on a given day. Factors like handle material or color, etc.

I'd guess that most people that read the magazine will understand M4 being able to dress 5 deer and S30V being able to dress 3 deer before a quick touch up is needed. I'd also guess that most of the people, even if they are a little interested, would not understand an article about the details of metalurgy. I think I would be more entertained by reading about the deer method than an accurate method for testing edge retention. Unless you can describe the material in a manner the reader will enjoy reading and comprehend, the article is useless even though it may be more accurate. I have nothing but respect for Cliff's knowledge about steel and knife edges but for me, and I'm only speaking for me, I don't understand what he is talking about most of the time. This is because of my lack of knowledge about steel and how different things effect the performance of an edge. So, if someone wants to sell me an article about edge retention of different steels I might buy the one involving cleaning deer after reading the first paragraph. After reading the first paragraph of the article Cliff wrote I'd probably put that magazine back on the shelf. :p No offense Cliff. :)

Just my thoughts which are as accurate as my edge retention testing methods. :)

Jack

Jack

Indeed Jack, it sounds like that article was intended primarily as entertainment.

jackknifeh wrote:Unless you can describe the material in a manner the reader will enjoy reading and comprehend, the article is useless even though it may be more accurate.

If you want very simple experiments, I can easily describe them and the results, here is a very simple test to rank the blade steels in the knives for edge retention :

-take each knife
-roll six normal dice
-add up the dice

The higher the score the better the edge retention of the knife.

--

Now this is where you might jump in and say "Wait, but that doesn't actually measure edge retention, it is just a made up list of numbers."

Exactly, and therefore you should not draw any conclusions from that list.

The article described has the same problem. The guy might have well just rolled dice and ranked the steels and he would have had an equally valid list of edge retention rankings. The reason this is the case is because the noise level is higher than the significance and thus what is produced is a random list.

If you want to see this in Calc (spreadsheet), as it isn't obvious then just do this :

-create in column A a list from 1 to 10
-in column B type in this forumla "A10*2+RAND()*100"

If you plot that then it will look random, but every number in column B is twice A, so it should look linear right? But it doesn't look like anything because while there is an underlying correlation it is masked by the random deviation.

There is actual work required to know something, if you don't do the work then you don't know, you are just making things up and you might as well save time and just roll the dice. If you want me to ignore this and pretend you can draw conclusions without doing the work, that isn't going to happen.

As an aside, I write the way I write because there are people who want to understand so they ask, if you don't want to understand, and all you want are the results then you can skip everything and just jump to the conclusions at the end.

However if you just want to read conclusions which are completely unfounded or claims made that are simply flat our wrong (though wonderfully simple) in that case I would suggest someone else, its a big world, plenty of room for everyone.

Tdog wrote:...
Gijoe, folks are just havin fun don't take this personal. You've got to admit, some of the posts here are pretty good...

+1

I'm just here for the articles, the pictures don't interest me that much

Cliff Stamp wrote:If you want very simple experiments, I can easily describe them and the results, here is a very simple test to rank the blade steels in the knives for edge retention :

-take each knife
-roll six normal dice
-add up the dice

The higher the score the better the edge retention of the knife.

--

Now this is where you might jump in and say "Wait, but that doesn't actually measure edge retention, it is just a made up list of numbers."

Exactly, and therefore you should not draw any conclusions from that list.

The article described has the same problem. The guy might have well just rolled dice and ranked the steels and he would have had an equally valid list of edge retention rankings. The reason this is the case is because the noise level is higher than the significance and thus what is produced is a random list.

If you want to see this in Calc (spreadsheet), as it isn't obvious then just do this :

-create in column A a list from 1 to 10
-in column B type in this forumla "A10*2+RAND()*100"

If you plot that then it will look random, but every number in column B is twice A, so it should look linear right? But it doesn't look like anything because while there is an underlying correlation it is masked by the random deviation.

There is actual work required to know something, if you don't do the work then you don't know, you are just making things up and you might as well save time and just roll the dice. If you want me to ignore this and pretend you can draw conclusions without doing the work, that isn't going to happen.

As an aside, I write the way I write because there are people who want to understand so they ask, if you don't want to understand, and all you want are the results then you can skip everything and just jump to the conclusions at the end.

However if you just want to read conclusions which are completely unfounded or claims made that are simply flat our wrong (though wonderfully simple) in that case I would suggest someone else, its a big world, plenty of room for everyone.

I agree. I used to wish you would give a bullet statement to a question. However, there may not be a bullet statement that could possibly answer the question. So I read your posts and get what I can. For example the calculus you mentioned in this post. I have no idea if it is accurate (I believe it is) or just a bunch of chatacters strung together. When you talk about steel at a level of knowledge I'm not even close to understanding I don't get anything from it. If I were to take classes or just use the inet to study about what you have learned from school, training, experience I would then understand you more. I don't have the energy to do that though. Mainly because there would be no hands on benefit from my efforts. When I mentioned I would buy the deer article and put yours back that's because of the entertainment value of the deer article. I appreciate your posts because I know there are those who do understand most if not all of what you write. One other thing I'd like to add. I have changed my opinion about a few things regarding sharpening based on your posts a year or so ago. Maybe longer. Even though I didn't understand the details of your theories I changed some things in my sharpening to find out if you were right or wrong. A lot of what you claimed proved to be correct IMO because I could see it in my sharpening results. It took at least a year of working on my sharpening ability to finally be able to see the improvement in the edge sharpness. More importantly, I'm getting better edge retention now than I was 2 years ago. I can see this in daily use when I pay attention to it. This is what I think even though I couldn't prove it in a million years because I don't have the desire to test my edges to be able to prove it. If I couldn't tell any difference in performance during daily use I wouldn't care if M4 holds an edge longer than VG-10. If M4 only out performed VG-10 under strict testing controls and then is only 6% better I don't know if the difference could be seen in every day use. If I had a job that required me to cut cardboard a lot and M4 stayed sharp till quitting time and VG-10 needed a touchup 1 hour before quitting time I'd consider those steels close enough to live with. But if a steel is significantly duller at lunch time then there is enough of a difference in performance for me to care about. I consider this example similar to the deer story. :) Or maybe my opinion of day to day performance is as accurate as my ability to guess a dice roll. :) Anyway, I'll keep reading your posts and getting what I can from them.

Jack

The Mastiff wrote:I read they had professional testers and would give each sword a grade, like "two bodies", or "1 body", up to 4 bodies. They used criminals, supposedly dead but there is some question about this. :)

I'm pretty sure that if they weren't entirely dead before the testing, they definitely were afterwards.

I'm pretty sure that if they weren't entirely dead before the testing, they definitely were afterwards.

Yes, I'd think so. After getting your torso cut diagonally from the shoulder/neck down through the ribcage and out the side the now separated body parts would have no blood pressure instantly. The diaphragm would be cut off from the nerves controlling it from the brain stem area, etc., etc.... Gravity would begin to do some work causing the organs to fall out.

Those rolled up mats were introduced as the best substitute for bodies after the testing was outlawed. If you have seen a test where the swordsman cut through two mats in one swipe that was a 2 body sword. Imagine 4 bodies. No wonder some of those knives were valued at estates and castles for the finest ones. Tameshigiri = test cutting from what I was told. I don't speak the language so I'm not sure about that.

The testers had to be pretty skilled and very consistent, which is the hallmark of true skill.