Welcome! Log In Create A New Profile

Advanced

Possible collaboration : Tom Hosang

Posted by CliffStamp 
Possible collaboration : Tom Hosang
June 01, 2020 10:15AM
Discussing now with Tom a possible evolution by him of some knives I will provide. For reference :

[www.youtube.com]

His data looks the most real that I have seen in awhile :

[docs.google.com]

-wish that had been blinded

[docs.google.com]

-note the large scatter, +25% from one run to the next

For data nerds out there, if your one shot variance is 25%, and you wanted to get that down to ~5%, you would need to average > 25 runs, which means if you wanted to compare just TWO knives and be able to tell a difference of ~10%, at a minimum you would need to to FIFTY runs.
KWB
Re: Possible collaboration : Tom Hosang
June 03, 2020 05:10AM
By that math the average user is going to simply use feelings as to whether the blade is good or bad.

Contact 570-486-9095
cKc
Re: Possible collaboration : Tom Hosang
June 03, 2020 06:35AM
Quote
KWB
By that math the average user is going to simply use feelings as to whether the blade is good or bad.

That is 100% accurate to what happens, which is why for the average user all this stuff about HT, steels etc is literally meaningless.

you give them a knife it either works good, sharpens easy or it doesn't..

now you give the avg person a steel that is unhardened, and they will struggle to sharpen it because of the burr, and it will dent a lot in rough use and they will call it a rubbish knife..

give the avg person a high carbide steel at 65+ rc and they might think its awesome at first and then it dulls, or chips and they struggle to sharpen it and nothing is happening, its taking forever and not getting sharp.. its basically as useless to the avg person as the un heat treated knife.

the avg person doesn't join forums on the internet and research all sorts of special techniques, and $1000 of stones for every occasion to deal with a special knife in a special steel..

the avg user wants something that cuts good for a while and rubs up sharp easy on a pull through sharpener, hone, or $2 stone.
those are the knives an avg person will say is a good knife in a good steel.

are then wrong? i dont think so. anymore than an F1 car is not useful by price or practicality for the avg car driver.

----------------------------------------------------------------------
It's not Cliff, its Dr Stamp
#kebabstickcut, it's a thing - make it happen
Re: Possible collaboration : Tom Hosang
June 03, 2020 09:51AM
Quote
cKc

give the avg person a high carbide steel at 65+ rc and they might think its awesome at first and then it dulls, or chips and they struggle to sharpen it and nothing is happening, its taking forever and not getting sharp.. its basically as useless to the avg person as the un heat treated knife.

A point that is often missed in many sharpening commentaries/rants is the common :

-gear
-time
-experience

A lot of these guys have, and even then, they can take ~30 mins to sharpen a tiny knife with minor damage. How many people send in Swiss Army Knives to sharpen, when was the last time you saw one of them in one of the videos? Is it because people who buy them can sharpen them because they are actual tool users and maintenance is a part of tool use, and as well due to the steel it simply requires little in terms of abrasives/skill.

The funny thing is, I was one of the first, along with Alvin and Mike, to push for high hardness steels in the 90's. I pushed Wilson on hardness, I did a lot of work showing the influence of hardness, and how drawing steels back to make them softer wasn't always the right choice. But it was just PART of the issue. It seems like lately all of the information was lost and just one random point was kept.

[www.youtube.com]

(explicit language, etc.)

This is a kind of commentary from a guy who makes knives (hobby maybe, he is a professional cook) commenting not only on another maker, but how that maker and another user didn't test knives in a proper manner and put at risk "a years worth of work" that him and some other guys did on the importance of hardness. However after watching a bunch of video's I still don't even know what was the point of the work they are doing, like what are they claiming.

Is it that as you increase hardness, with a given gross micro-structure that you will see :

-an increase in strength
-increase in wear

which in many types of cutting leads to an increase in edge retention? Well that is obvious from the materials data, you can just look that up. And if you want to know the types of gains :



That's from ~7 years ago, a fully blinded run, showing a linear increase in edge retention with hardness, and due to general variances in hand cutting, IF, you are willing to do a bunch of runs, you likely could see a difference if the hardness was > 2 HRC points. If the hardness difference is really large, > 4 HRC, you will likely see it with less runs, less care, and if the HRC difference is very large > 6-8 points the performance maybe doubles so you could see that likely without too much trouble.

I was not even the first person to do that, that information has been around a long time. Mike and Alvin were doing it in the 90's, and there were people before him as well. It is the internet age, this information is so easy to find.

This has been done for a long time, Steve on BladeForums did one of the most meaningful trials and he found that in order for him to really see an obvious difference, he had to use 420HC vs S90V (exact same blade, exact same maker). Even then he wasn't looking at horrible vs awesome, he was just looking at a difference he could see without very careful measurement. And if you look at the actual materials data, that is about what you would expect, a big change in composition will give you ~double performance.

But if you think about real use, there is so much variation in what you cut, how you cut it on a day to day basis, even then you are only likely to really see it when you look at average sharpening/maintenance times over weeks/months. A professional who sends his knives out to be sharpened - likely could argue for a real difference, someone who does it as a hobby, harder argument, and most of what you read is really hyberbolic.

I have lived in third world countries where people make knives out of mild unhardened steel and they use them all day with minimal sharpening. Steel is so much harder and wear resistant than what it cuts, it wears/deforms slowly. You have to remember steel is only recent, and knives (and swords) were made out of much softer materials for a LONG time.

The curious thing to me is like this is what seems to happen :

-people get excited, do a bunch of work
-new people come in, they do NO research and start from the beginning and "discover" it again

And it seems like the longer this goes, the lower the quality of the information as discussion and collaboration, especially criticism is just lost.

As an experiment, in the last two weeks I made > 20 comments on a number of these channels, all on people who constantly talk about how serious they are about information. Almost none of them got a response. If you wan to learn, the critcism is where you do it, praise does nothing except feed your ego.
cKc
Re: Possible collaboration : Tom Hosang
June 03, 2020 10:13AM
>>This is a kind of commentary from a guy who makes knives (hobby maybe, he is a professional cook)

from what I gather, he is not a knife maker. he made 2 test knives to prove a point. and apart from that, he is just an enthusiast from what I can see.

Apart from a rare number, it is all too common that the knives claimed to be the best in breed, are never shown to be best in breed because the prices are so high, people baby them, and pick up a cheaper option to do the work.

there is always the exception to the rule of course, and i can only cherry pick this information from what i see on the internet, and the statements from almost every single customer or friend of mine in general buying this stuff and then being too scared to use most of it for a fear of investment loss.

unless you can start getting the "high" end as they are called steels at a price point where people that rely on knives for their work can afford to give them a spin and get real feedback its hard to say

----------------------------------------------------------------------
It's not Cliff, its Dr Stamp
#kebabstickcut, it's a thing - make it happen
Re: Possible collaboration : Tom Hosang
June 03, 2020 10:35AM
If you wanted to actually see them in real use, one possible experiment would be to make utility blades from them and give them to tradespeople. The only restriction would be on stock thickness as those blades are really thin as they are essentially heavy razor blades.

The other thing would be to make a very simple, and very small belt knife pattern with a very basic set up and design. Joe Calton and Jeremy McCullen make lots of patterns, their paring knives in particular.

You could get away with very little steel, ~1/16-1/32" thick, < 0.5" wide, 4" in total length (blade and partial tang).

Around here, a lot of guys using paring knives as utility knives :

-cut very well
-ergonomic
-free (they steal them from the kitchen)

You can buy decent ones for ~$5, just get a full flat grind on thin stock. They often even give them away on promotions when new lines of knives come out. Each time that happens you see them everywhere on work sites.

However, the results from this you might not like as these guys :

-cut actual used/dirty material
-do a lot of scraping
-cut on dirt/concrete
-cut very harsh materials like fibreglass insulation, shingles, etc.

They normally just use the knives until they don't cut and get a new one. I used to pick them up and refinish them all the time :

-grind a new tip
-grind the bevel back on
-flatten the primary back to 0.005" at the edge

This would take < 5 mins on a 1" belt sander with a low grit belt. By hand it would not take that much longer with the proper stones, maybe 10-15 mins.
Re: Possible collaboration : Tom Hosang
June 03, 2020 11:11AM
cKc, Cliff,
See, that is why I like talking to you guys. You can explain, in detail, and in practical terms, the smaller details while including the whole picture. I can only focus on one or the other, and although I can bridge the gap between the two in my mind, I have a hard time relating the two to others, even in a simple manner. Both of you are able to do that and go in depth. Which is awesome and much needed. Often, I have found that once I explain a topic to someone, and I think they understand, when I talk to them a week or two later, they are back stressing about high carbide steels and edge retention, like nothing else matters. Those are literally the last two things I care about when picking a knife for the kitchen, outdoors, or EDC. We got by for literally THOUSANDS of YEARS using simple steels at less that 55HRC. Right now, even though people are getting interested in high HRC knives and cutting performance, there focus in limited and doesn't reflect actual knife usage. This is a HUGE problem for me, because if it isn't practical, then what are we talking about? I like getting edges that will pass a HHT 3/4 right off the stones, no stropping. Yet, other than to tell me I did my part on the stones when it came to sharpening, it doesn't reflect in actual use. An edge that will tree-top or even just catch hairs above the skin cuts basically the same in what I cut for EDC, I cannot tell the differnce. I can do that with $15 dollars worth of stones in under 5 min. Yet so much is made of things that acutally do not matter, it is all gearhead stuff. I am a gearhead also, but I do like using my things and if not keeping my skills up to par, improving on them at least. But I rarely find anyone else, save the guys here on the forum, and a few custom makers and knife sharpeners, that feel the same. It's like we have lost the plot.

------------------------------------------------------------------------------------------------------------------------------------------------------
"Gotta love living in 2019 baby, (63rc too soft on a production knife)"
--Shawn Houston
__
"I am still discussing issues of steels and performance at this stage."
--Cliff Stamp
__
"Cause geometry cuts, .....steel determines the level and the duration"
--Roman Landes
__
"But in general, I'm all about high performance, Ergos, safety. That's why I've been accused of 'designing in the dark' "
--Sal Glesser
------------------------------------------------------------------------------------------------------------------------------------------------------
StoneSharpEdges

Jason D. Stone on YouTube
Re: Possible collaboration : Tom Hosang
June 03, 2020 01:43PM
Another source of conflict is science. I really don't think most people understand what science is, what it is used for, it is capable of, and it's limitations. Yet I see things either claimed or detracted in science's name, to either prove or disprove whatever point of view is being argued. This is very unfortunate, because science is one of our greatest tools in understanding and interacting with the world.

------------------------------------------------------------------------------------------------------------------------------------------------------
"Gotta love living in 2019 baby, (63rc too soft on a production knife)"
--Shawn Houston
__
"I am still discussing issues of steels and performance at this stage."
--Cliff Stamp
__
"Cause geometry cuts, .....steel determines the level and the duration"
--Roman Landes
__
"But in general, I'm all about high performance, Ergos, safety. That's why I've been accused of 'designing in the dark' "
--Sal Glesser
------------------------------------------------------------------------------------------------------------------------------------------------------
StoneSharpEdges

Jason D. Stone on YouTube
Re: Possible collaboration : Tom Hosang
June 03, 2020 07:43PM
It is odd to me because science is like the simplest thing in the world to understand. Science is a way of knowing, specifically it is knowing by interacting with the physical world. This is in contrast to say Math which is knowing by seeing what can be deduced from given assumptions, following given rules of inference.

That is what is kind of amusing when people say things like "ok, this isn't scientific but ..." and then proceed to make very strong claims. Listen, if you say it isn't scientific then by definition this means you are saying you don't know what you are talking about. It literally means you did a bunch of work and you learned nothing from it.

[www.instagram.com]

Now you can do precise measurements, and not be scientific, for example in that link there is a bunch of data, however I have concerns about the conclusions reached which is why I commented :

Quote

"As a point of concern, M4 is a HSS, it is designed to actually retain working hardness when it is hot (as in glowing), that is the entire purpose of all of that secondary hardening alloy, hence why drill bits and such are made from it.
If you look at the tempering curve, drawing the hardness on that back to 53 would require serious temperatures, especially since it is back from the edge significantly and still very soft as you said so the actual body of that blade had to get > 1100 degrees. In the videos I have seen, they are holding on to the blades, which would be difficult.

However, I mean they could do a really shallow austenization, not put significant alloy in solution and that will really compromise the resistance to softening, but it also would stop the blade itself from hardening.

Was the blade ever returned to Bark River for comment? Does the edge behave as it if is almost 50 HRC, like that is a huge different in hardness, that is softer that almost all production knives so it should be easy to physically deform it.

If it behaves that hard I would be curious to know what they did to get that and what they said about the results."

In order to be scientific (i.e. to know that the blade is soft) :

-test a blade of similar shape in the same way to make sure that you are not seeing distortions from geometry
-if the blade is ~50 HRC at the edge it should deform readily compared to a 60 HRC edge, do a physical test
-figure out what kind of HT could actually cause a blade to draw down when it is a HSS (ask someone if you can't do this)
-contact Bark River and see what they have to say

Unless you do that, then I don't think you know that the blade really is soft. If you do the first two things I think you are in a decent position to say you know the blade is soft, then is the curious part, as looking at the tempering data for M4 :

[www.hudsontoolsteel.com]

Just look at how hot you have to get it for it to soften, and keep in mind the steel will be visibly red by then, and you are somehow holding onto the blade while it is glowing red at > 1100 F. All of that makes me very hesitant to say it is soft, even if I did those measurements. There are just so many questions which scream out to be answered.

The thing is, that guy is a knife Maker, and he is casually saying that another maker/manufacturer is making defect products (if you look at his comments, he claims this is the norm from what he has seen). This kind of thing is really unusual in industry in general because of liability concerns, because it is really easy to show that conclusions are not supported by data - hence, well you are liable for damages.
Re: Possible collaboration : Tom Hosang
June 03, 2020 08:28PM
Cliff,
Exactly. I don't understand why people don't understand it. When I bring up science to some people, they start talking about scientism, which isn't what I am talking about when I talk about science. If you ask me, the public educational system is broken, since they can't even teach the fundamentals of science to the mass of peoples, even at a basic level. This hurts me greatly.

------------------------------------------------------------------------------------------------------------------------------------------------------
"Gotta love living in 2019 baby, (63rc too soft on a production knife)"
--Shawn Houston
__
"I am still discussing issues of steels and performance at this stage."
--Cliff Stamp
__
"Cause geometry cuts, .....steel determines the level and the duration"
--Roman Landes
__
"But in general, I'm all about high performance, Ergos, safety. That's why I've been accused of 'designing in the dark' "
--Sal Glesser
------------------------------------------------------------------------------------------------------------------------------------------------------
StoneSharpEdges

Jason D. Stone on YouTube
Re: Possible collaboration : Tom Hosang
June 03, 2020 09:09PM
In general, when people complain about scientism, it is almost always the case that they don't understand science. There are lots of questions that science can't answer, but generally in regards to knives, well it can.

It is as simple as this :

-are you trying to describe what happens to physical objects and why it happens

Well that is what science does, and again this is the key point to keep in mind in order for something to be scientific :

-are the claims you make supported by the data

If so, then it is science, if not - well it isn't. Hence be careful about what you claim, that's it.

Can you make statements about steels and heat treatment without fancy equipment? Of course, there is no requirement in science for specific equipment, it isn't like science is a recent thing, there was science long before the digital age, and even before there were mechanical measuring devices. Again, you just have to make sure that what you say is supported by the data, bias is the main issue. Most people don't know what real data should look like, hence you can almost always spot someone who is not really measuring, but almost creating data because it looks artificial :



Real data looks like that. You can see a clear trend, but it has wonky behavior at times because well there are always random and systematic sources of error. The Fiskar's data kind of looks like my skill went down, but that's just random variance. The Fiskar's group is actually the second highest group and it consistently high, ~91%, but just due to random chance, it looks like it went down. If you generate random numbers, you will see patterns at times which is why you often want LARGE data groups to prevent false conclusions.
Re: Possible collaboration : Tom Hosang
June 03, 2020 10:42PM
Cliff,
Exactly. Like you pointed out before, people are doing work and prefacing it by saying 'this isn't scientific' when they should be saying something else. Like there is no bias control, only accurate to a certain specification, and margin of error. People want to substitute 'real world testing' to use instead of what they consider 'science' or 'scientific'. There is no reason why that type of work can't be considered scientific, as long as a few things are followed. I never understood the difference between 'real world' and 'scientific'. They should be the same thing, right?

------------------------------------------------------------------------------------------------------------------------------------------------------
"Gotta love living in 2019 baby, (63rc too soft on a production knife)"
--Shawn Houston
__
"I am still discussing issues of steels and performance at this stage."
--Cliff Stamp
__
"Cause geometry cuts, .....steel determines the level and the duration"
--Roman Landes
__
"But in general, I'm all about high performance, Ergos, safety. That's why I've been accused of 'designing in the dark' "
--Sal Glesser
------------------------------------------------------------------------------------------------------------------------------------------------------
StoneSharpEdges

Jason D. Stone on YouTube
cKc
Re: Possible collaboration : Tom Hosang
June 04, 2020 12:03AM
Quote
Jason
right now, even though people are getting interested in high HRC knives and cutting performance, there focus in limited and doesn't reflect actual knife usage. This is a HUGE problem for me, because if it isn't practical, then what are we talking about?

i think it is good to push boundaries, and explore what things can and cannot do and see what the min and max limitations are of steels and HT and everything. its all good for learning.
but when you start taking a very isolated concept and pushing it as the ultimate, and specialty, and everything else is inferior without determining if your perspective is myopic then things start getting crazy.


imagine for a moment that we were talking about shoes, and this was an entire discussion about wear and longevity on the soles of the shoe,

we start looking at cobbling a shoe with ceramic plates.. high wear, not very tough.. but we put them on our shoes and claim to the world, these soles have the longest wear resistance anywhere and this is going to provide the best shoe on earth "Period"..

now we demonstrate our shoe by getting a boat shoe with a soft natural rubber sole, and we walk back and forward on our driveway carefully for 2km, back and forwards in 20m strips, and show the rubber wearing down..

now we get our ceramic soled boot, and we tap it lightly on the concrete and say, look, its not brittle, nothing is shattering.. its as tough as i need, and we walk 10km carefully back and forward on the little walkway, and show, how impressive the wear is on this sole.. its a harder sole, it has less wear, its better.. "period" and all those people out there that disagree dont know squat..

Great..
so now some guy buys a pair of boat shoes in ceramic soles because this is the best stuff. it has the longest wear.. he walks around the boat timidly, skating all over the place and slipping around.. steps on a clinch pin by accident and snaps the sole in 1/2 leaving his foot bleeding where a piece just dug into him...
not so great in real work applications.

"well what an idiot, he clearly abused that shoe" bla bla...


lets not even get into the running shoes.

----------------------------------------------------------------------
It's not Cliff, its Dr Stamp
#kebabstickcut, it's a thing - make it happen
Re: Possible collaboration : Tom Hosang
June 04, 2020 08:48AM
Quote
jasonstone20
I never understood the difference between 'real world' and 'scientific'. They should be the same thing, right?

I have worked in many labs, as far as I know, they took place in the real world.

To be frank, I think those "this isn't scientific" claims are just there for people to avoid having to be responsible for what they say. It allows them to make very strong claims, promote a patreon, openly talk about inferior/superior brands/steels, call out makers for ripping off customers - and then if you make a criticism they retreate to "this isn't scientific".

For the last two weeks I have been making the odd post on YT and now even Instagram, and the engagement is very low with people who want to look at how they are doing things to make sure they are drawing well supported conclusions. But that doesn't stop them from making very strong claims on a constant basis.

Ironically, they are right, what they are doing surely isn't scientific, because making claims and not being willing to take peer review, well there is nothing more unscientific than that.
Re: Possible collaboration : Tom Hosang
June 04, 2020 09:24AM
Cliff,
Actually, what I should of said about science and scientism is this:
When talking to people about science, and they start criticizing science, what they are actually criticizing scientism, which they mistake for science.

------------------------------------------------------------------------------------------------------------------------------------------------------
"Gotta love living in 2019 baby, (63rc too soft on a production knife)"
--Shawn Houston
__
"I am still discussing issues of steels and performance at this stage."
--Cliff Stamp
__
"Cause geometry cuts, .....steel determines the level and the duration"
--Roman Landes
__
"But in general, I'm all about high performance, Ergos, safety. That's why I've been accused of 'designing in the dark' "
--Sal Glesser
------------------------------------------------------------------------------------------------------------------------------------------------------
StoneSharpEdges

Jason D. Stone on YouTube
Re: Possible collaboration : Tom Hosang
June 06, 2020 10:38PM
Quote
CliffStamp
For data nerds out there, if your one shot variance is 25%, and you wanted to get that down to ~5%, you would need to average > 25 runs, which means if you wanted to compare just TWO knives and be able to tell a difference of ~10%, at a minimum you would need to to FIFTY runs.

Quick question, how did you go about this calculation?
Re: Possible collaboration : Tom Hosang
June 06, 2020 10:52PM
Quote
mindsmirror


Quick question, how did you go about this calculation?

In general, with some assumptions about data which are usually valid, if you do n runs, you will reduce the spread by sqrt(n).

So if say your spread in data is 10, and you do 25 runs and you average that, your spread in averages will be 10/sqrt(25) = 2.

The assumptions are :

-the data is normally distributed or close to it
-your sample is fairly big
-you don't hit outliers

If you have data prone to outliers you are better off with medians and median of the absolute deviations which is what I normally use for chopping (as wood is insanely variant) and for cardboard as well.
Re: Possible collaboration : Tom Hosang
June 06, 2020 11:18PM
Quote
CliffStamp
Quote
mindsmirror


Quick question, how did you go about this calculation?

In general, with some assumptions about data which are usually valid, if you do n runs, you will reduce the spread by sqrt(n).

So if say your spread in data is 10, and you do 25 runs and you average that, your spread in averages will be 10/sqrt(25) = 2.

The assumptions are :

-the data is normally distributed or close to it
-your sample is fairly big
-you don't hit outliers

If you have data prone to outliers you are better off with medians and median of the absolute deviations which is what I normally use for chopping (as wood is insanely variant) and for cardboard as well.

Ah, so basically standard error. Thank you for the info!
Sorry, only registered users may post in this forum.

Click here to login