Hacker News Reader

Comment by hprotagonist | Hacker News Reader

pattern welding is, effectively, a solution to a problem we don't have any more.

When steel quality varied widely between batches, you needed a reliable way to mix it up so that the overall properties of the resulting object were predictable. This evolved into a more sophisticated technique that gave you not only predictability but differential properties (harder steel at the edge, more ductile in the spine) and a pretty visual result.

The batch variability problem isn't a problem any more, and we can get differential properties with very precise heat treatments now, so all that's left is the pretty visual result.

So, if it appeals to you; great! Just don't think you're getting anything that's super magical above and beyond what a modern alloy can do.

View on HN

I know this does not have anithing to do with knives but Samurai swords are made from four different kinds of steel. The core, back, sides and the cutting edge. The sword is straight when forged and the arc comes from the hardening when the cutting edge expands more than the rest. Source: witnessed in person the first time when a katana was forged outside of japan at a festival at the japanese culture institute in cologne. The swordsmiths have the rank of a priest and wear white robes.

From a pure material perpective, the issues, also for the Japanese, was poor steel and iron ore quality (as an other comment stated above). So in order to get all you inpurities out of the final product, melting processes weren't up to the task yet, you are basically obliged to pattern weld the shit out of your steel.

Particularly in Japan, this was compounded by having iron ore mostly as part of a sand-ore mixture (without magnets getting the pure ore out without mixing it with sand is close to impossible). As a result the Japanese came up with a very smart forging technioque to solve that problem. And the swords, at least the high end ones, are pieces of art. Not fielding full-plate armor like the Europeans did, cutting remained a desired quality for a soldiers side arm. So they stuck with it. Simultaniously, the Europeans developed spring steel and weapons developed in a totally different direction.

I’m not sure that is fully true - the authority on this that I think has proven out his trustworthiness in the knifemaking community, Kevin Cashen, makes a compelling argument that with correct heat treatment, pattern welding gives you the ability to mix alloys. He frequently uses an O1/L6 blend.

I am not saying that most of it isn’t cosmetic - that is certaninly the point most of the time, but, for boutique knife makers, it can give you more options if what you want is not available in a commercially made alloy. Keep in mind that most commercial steels are not the “best possible” for knives, but frequently the best balance of performance, and manufacturability (I.e. they avoid very complicated heat treats, and very expensive hard grinding with lots of embedded carbides).

To my understanding Damascus was indeed not needed anymore as sword fighting came out of fashion. For swords it was very useful to have flexible yet sharp blades. Or has Damascus steel been traditionally used for knives too?

The idea is that Damascus steel was a technology from before there was much metallurgy knowledge.

These days we have a molecular knowledge of steel and can control and test the microstructure at that scale. If you really want the best knife you look to metallurgists not artisans. Whatever you could do with Damascus steel you could do better with modern alloys and science.

There's nothing wrong with the old ways, but there is a lot of magical thinking about knives and swords that doesn't have much basis in reality. Buying in to the mythology can be fun as long as you don't get too attached to the myth and start confusing it with actual utility.

In other words, nobody owns a Rolex because they think it'll do a better job at letting them know the time.

For swords, I gather that key features are 1) notching or shattering other blades, and 2) resisting being notched or shattered by them.

Are you arguing that modern alloys and methods can do a better job at that than Damascus or other classical methods involving layering and folding?

That is the case, yes.

Cool. I guess it's just that there's a lot more demand for gun barrels than swords in recent centuries. So there's not much practical experience, and the delusion persists that the old ways are better.

But still, the patterns can be beautiful. I've seen blades with patterns that resemble engravings. Maybe the approaches could be merged, to yield modern performance with classic beauty.

But why? The hypothetical people who would actually fight with such things today don't care about how it looks. Go to a gun show or AR15 enthusiast forum and compare how those guns look with those that people in the military or SWAT teams use. Playing with guns/swords is fundamentally different from actually relying on them not to die. And then you have the people who like knives/swords for how they look, and would like to be able to argue that 'their' sword/knife also has the most utility if it were to be actually used. But they already do that, and are fully convinced of it, too! And the lack of actual substantive evidence (even if one could devise a single optimum function by which to measure each knife/sword...) is what keeps all these people, and the whole industry, going. So it's in nobodies interest to have one single 'best' thing. (and that's even ignoring that the amount of idiocy and stubbornness in such communities is staggering - look at the comments on the OP!)

Yeah, I guess. Me, I just have a military-issue machete. It isn't pretty, but it's tough, and holds a good edge. All I've killed with it are weed trees, though ;)

Yes, but it's also difficult to say there is strictly a modern/classical dichotomy and which method or material would belong in which bin.

Someone could certainly do some studies comparing methods and results, but anything of that matter would be very easy to bias.

What is clear is that classical methods are being employed by artists and not scientists. Mythology and visual appeal are being used to sell things, not quantifiable facts.

Yes, I get that.

What matters now is mostly appearance. And maybe vague stuff about ability to keep an edge. But back in the day, you were in deep shit if your sword shattered. Or even got notched, if that put you off the flow.

What scenario am I in where I would have a sword, have it shattered and be in trouble?

In war, people mostly use polearms. In a sword fight, as has been shown in many HEMA videos i have seen, it's over in a single exchange. At that point it doesn't matter if your sword is shattered or not. And the idea that your sword being notched will 'put you off the flow' just doesn't buy it for me.

Basically whenever you are using a sword that isn't some ritualized combat.

If you are using it in a battle and your sword breaks after the first exchange, what then?

Here is an hour long documentary answering your question: https://www.pbs.org/video/nova-secrets-viking-sword/

> For swords, I gather that key features are 1) notching or shattering other blades, and 2) resisting being notched or shattered by them.

You're thinking of theatrical fencing. All that cling-clang-cling makes for great visuals and sound but is undesirable in real combat.

Unless you have two knights in full armor clobbering each other over the head (at which point a mace is probably more efficient than a delicate sword) sword-on-sword contact is a means of last resort when you can't move out of the way.

Although D&D may have taught you differently, a "sword block" is a wasted counter-attack. Why bother binding your sword if you could have used the opportunity to slip out of the way and catch the opponent off balance and end him?

In fact, blocking is what sidearms are for. Grab a dagger or a buckler and take down the opponent's sword without binding yours. If swords do touch in combat it's usually at an angle and just enough to deflect the blow without overcommitting too much to risk wasting a the chance to get in a counter.

(full disclosure: I'm talking about European medieval swords and from having watched experienced HEMA fighters)

Those are good points. From Stephenson's Mongoliad, I get that binding the opponent's sword, and closing, are key moves. But I'm most familiar with Japanese style, from studying Aikido. We sometimes practiced with wooden bokken. And yes, any blade-blade contact is glancing.

But still, stuff happens.

Yeah, I think point 2 is still relevant, but intentionally damaging the opponent's sword is pointless. Ideally the fight doesn't last long enough for the accumulated damage to matter and focusing on damaging the sword is a worse strategy than focusing on incapacitating the person holding the sword (who by the way might still be dangerous without the sword).

Well, there is the move of capturing the opponent's blade against the guard, and then closing to stab with one's knife. The risk there is that your guard will fail. But if the opponent's blade breaks, they're likely in worse trouble.

> Although D&D may have taught you differently, a "sword block" is a wasted counter-attack. Why bother binding your sword if you could have used the opportunity to slip out of the way and catch the opponent off balance and end him?

I dunno, I did a bit of fencing and I know there are all sorts of ways to bind the blade and counterattack simultaneously.

In fact, the a HUGE portion of the Die Blume des Kampfes (historical weapons manual) clearly includes a variety of sword-binding techniques. See the pictures yourself if you don't believe me:

http://www.wiktenauer.com/wiki/Die_Blume_des_Kampfes#Sword

It would seem that binding, blocking, and counter-attacking is a fundamental skill of historical sword fighters.

In my experience, it is virtually impossible to avoid the opponent's blade like a Hollywood sword master. Swords are surprisingly light when held with two hands (even a 5lb Claymore moves very quickly, even in the hands of a weakling!!), far faster than you can move your torso or even legs.

EDIT: Lots of other stuff here, but I was ranting. So I deleted it.

Sure, but how many of those techniques involve intentionally damaging the opponent's sword? The emphasis in what you said lies on "and counterattack simultaneously". If you overcommit into a block enough to damage the opponent's weapon, you won't be able to counterattack.

The reason the 5lb sword still moves swiftly is that it's not used for a Hollywood cleave.

I agree with you 100% that Hollywood-style direct blade-blade blocks are rare. At least, for skilled fighters. But they do happen, as a last resort. You get knocked down, say, and use your blade as a shield.

But about those Claymores. You're right that they'd be slow for cleaves, with the huge rotational inertia. But in melee, it's my impression (perhaps colored by too much fiction) that one can make that an advantage. Basically, by never reversing direction, but rather by following 3D Lissajous curves. Which isn't really that different from Aikido.

> following 3D Lissajous curves

The only problem with that is that it creates predictability, which is what you want to avoid.

That said, my entire argument was that it's not helpful to imagine swords as having been specifically designed for causing damage to other swords or even withstanding repeat blade-on-blade strikes for extended period of times. And we seem to agree on that.

Actually some people do go for Rolex because +/- 2 spd guarantees much less of a hassle than certain other mechanical watches' upper bound of +5, +6, or +10 spd.

If someone wants to know the exact time, they don't get a mechanical watch. Everything else is post hoc justification.

Cheap Casio has better accuracy than any mechanical watches.

This is not actually true anymore, but it might have been true a few years ago.

https://www.digikey.com/product-detail/en/citizen-finedevice...

http://cfd.citizen.co.jp/english/prod-tech/product/pdf/datas...

This $0.79 (yes, 79 cents) quartz crystal is specified at 5 ppm (parts per million), or 0.432 spd.

That's its manufacturing tolerance by the way. With proper calibration, you can get it better than that. (Ex: a microcontroller tuned against an atomic clock, you can add / subtract cycles to get higher accuracy. Like, every million cycles, add or subtract a 5 cycle offset to correct for any manufacturing error).

The temperature turnover point is guaranteed to be 25C +/- 5C. Which means the crystal won't drift due to temperature somewhere in the 20C to 30C range.

Quartz crystal clocks from cheap Casio watches are basically all you need with modern technology. Low cost, low-drift, super cheap, high-reliability from manufacturing, cheap electronics (just a counter + a cycle offset if you calibrate), cheap calibration (atomic clock radio signals from any 1st world country is free).

32768 is the perfect number for 15-bit register. Move the second hand every 2^15 cycles (aka: every 15-bit overflow), possibly with a bit of offset if you have a calibration routine somewhere.

I specifically specified mechanical watches. You can even match plain quartz oscillators with new exotic mechanical escapements, such as one from Zenith Defy Lab.

You can get temperature compensated quartz watches that run at 10 or 5 spy. Note that watches spend a lot of time above 30C. There's not much point in calibrating relative to an atomic signal in a watch unless you're already thermocompensated.

It's arguable there's also no point unless you have hand correction tech (seen in some Casios and Citizens) too.

Reply on news.ycombinator.com