Hacker News Reader

> A factory can ensure precision and consistency at a level that those hand crafting things never could.

Idk, in the example "a built racecar engine" one could imagine that instead of running CNC machines all day long to make pistons, replacing tools for wear periodically, a machinist could take a brand new end mill, make some initial test cuts and validate the dimensions, then make the final part and look it all over with the CMM or other tools. You could, by paying extra attention to each part, make a more precise part than something being made on an assembly line where things are not double checked quite so meticulously, for basic efficiency reasons.

Like I assume that by the same definition of "hand made" as the racecar engine, we would consider NASA rovers to be hand made. And they are made to a very high precision. We could consider whether SpaceX Engines have ever been "hand made" and how the quality of those relates to production engines (tho in this case, production engines must have rigorous validation of basically every component I would imagine, so the factory perhaps meets or could even exceed hand production runs in quality).

But I would not say it needs to be a hard and fast rule.

I suspect the original author’s use of “a built racing engine” was a bad example since there is no factory line for such things as they are too low volume. However, if we loosen the definition of racing engine to include hot rods, then I believe proven factory built designs are highly valued. To give one example:

> The 2JZ-GTE, which sits in the heart of the Supra, is one of the best inline-six out of Japan and easily tunable to 500+ Hp on stock internals.

https://www.drifted.com/1gz-fe/

here is the website that proves your statement as incorrect, Ford will even sell you a track ready mustang if you want it.

https://performanceparts.ford.com/engines/

You might have had me convinced, but then you mentioned a Mustang, which is not a race car. Now I am skeptical.

Are any of these engines used in NASCAR, F1 or any professional autosports? If not, they are not “racing engines”.

By the way, for very low volume parts, it would make sense to make them by hand, so just because Ford lists them does not mean that they are made on an factory assembly line. That is if they are actual engines used in professional autosports. If they are hot rod engines, a factory production line would make sense since those have a bigger market.

It seems like you're comparing a tight-tolerance handmade part with a loose-tolerance mass-produced part. But it's possible to mass-produce parts to any desired level of tolerance. It just costs more. TSMC is currently mass-producing parts with tolerances on the order of a nanometer.

It also costs more to handmake parts with tight tolerances than to handmake them with loose tolerances. So from a certain point of view tolerances are all about your budget.

Here's the crucial part, though. You can get any given tolerance level more cheaply by handmaking things at low volumes, and more cheaply through mass production at high volumes.

But let's talk about consistency. In today's statistical process control regimes, it's common for mass-produced parts to hit "six sigma": the point where the tolerances they're manufactured to are six standard deviations out from the process mean, so the normal variation in the process will only produce out-of-tolerance parts in one of a billion cases (1 - pnorm(6) in R). In practice, uncharacterized sources of variation that don't follow the normal distribution tend to overwhelm that, but it's still common to have a better than 99% chance that the part the manufacturer sends you will meet its specifications.

By contrast, when you contract with someone to make the first instance of something, if it's a very reputable maker, you might get 95%. The other 5% includes things like misunderstandings of the specs, unforeseen accidents, and undetected sources of deviation. In complex fields like architecture and software your chances of getting what you ordered are more like 0%.

I'd argue it's about customisation, not as much about the process. A couple can't be made anywhere other than a factory, but we're seeing this exact thing happen more and more with custom chip design. Sure not hand made, but skewing towards custom. Previously, there were a handful of CPU's and that was it, but then GPUs came along, and now Tensor, Graviton, etc all with more specialized purpose. Maybe it won't be too long before people start building fabs in their basement, but I doubt it.

Also, custom PCB's are getting easier and easier, it seems like there's a ton of YouTube videos sponsored by and about doing custom PCB's for projects in 2024. Heck, I have an Air gradient in my house, and while it's not a custom CPU, it is a custom, relatively low volume PCB designed by (as far as I can tell) a single person, almost like an artist...

I'm not trying to add a lot to the conversation, just suggest that even chips are moving towards customisation and artisanship.

We design our own PCBs at work. They are still made on a factory assembly line. The assembly line is just more flexible to be able to accommodate various designs.

That said, I think you have a good perspective and it was refreshing to read.

High end automated processes can indeed hit extremely high tolerances, but that’s a trade-off on cost (give or take) and often not one the factory is making. I’m reminded of a story about GM, where the top execs would preview drive all the new models - except the ones that were given to the execs had been hand-tuned by engineers to fix the manufacturing slop, so the GM brass was convinced their cars were much better than they were.

I am reminded of the following story:

https://www.1factory.com/quality-academy/forgotten-lessons.h...

Ford had transmissions produced in both the US and Japan, yet the Japanese made transmissions had 1/4 the repair frequency. Their engineers disassembled them to discover why. It turned out that they were both within spec, but the Japanese transmissions had been manufactured to tighter tolerances and had targeted ranges in the specification that resulted in a better fit and superior performance.

In this case, the factory had taken it upon itself to improve the design within the tolerances that Ford had specified. Tighter tolerances usually cost more, although economics of scale can bring the price down. Presumably that happened in the case of the Japanese factory. Otherwise, they would not have been able to produce the transmissions at the prices Ford paid them.

Disappointed the story didn't end with the revelation that the good transmissions were all hand made by monks.

Eliminating sources of variation from your manufacturing processes can reduce downstream costs in a lot of ways. The most obvious one is less rework, but also you need, for example, less inventory to buffer variations in production pace. Consequently it's now conventional wisdom that, although hitting tighter tolerances costs more, doing what the Japanese were doing normally costs less.

In regards to objects like the guitar listed in the quote, there is a draw to how a hand-crafted item can be _unique_ due to slight imperfections. This can range from just slight imperfections in the wood to things such as believing your guitar has a unique sound.

Sure the factory guitar may sound "perfect", but the only guitar that sounds like this handmade guitar is this handmade guitar.

I would not consider that evidence for the items being better, but it does explain why some people are drawn to hand made items.

I consider such items to be worse and not just because they are objectively worse. Items break and are lost. When you become emotionally attached to an object such as a unique imperfect guitar, you are guaranteed to have sorrow when it is lost (e.g. in a fire).

In any case, the original premise was that such items are better and thus hand made medicine is better too. However, you do not want variations in medicine. During the pandemic, my doctor had a suspicion that the dosages of certain drugs had been lowered by the manufacturers versus what was on the label. That posed a problem for him because he could not properly treat patients if he had no clue what was actually in the pills filling his prescriptions. His suspicion had stemmed from seeing a trend among his patients where the efficacy was dropping in a way consistent with the dosages being lowered, although he was not sure by how much. If he had tried to compensate by prescribing larger dosages, when the manufacturers began to get things right again, his patients would have been taking overdoses. His solution was to switch them to alternative drugs and hope that the manufacturers of those drugs were producing pills that matched their labels. This is why precision is very important when it comes to medicine and the last thing anyone needs are imperfections in drug manufacturing.

> Items break and are lost. When you become emotionally attached to an object such as a unique imperfect guitar, you are guaranteed to have sorrow when it is lost (e.g. in a fire).

Life is loss. You can’t have sorrow without having experienced joy. Sorrow passes.

You can avoid this kind of loss by getting a guitar made from a long running factory line such that you can get an identical replacement.

You can avoid a lot of joy that way, too.

I do not see how. A well made guitar from a factory would give the same joy as a hand made one as far as I am concerned. Perhaps even more upon realizing imperfections were avoided.

You two are in a battle over aesthetics, it is a battle that neither of you can win.

There is something attractive about a new set of work gloves. They are fresh, clean, almost begging you to use them in a project.

There is something attractive about a set of work gloves that have curled to the shape of your hands and have been burnished by sap and sawdusts and oils: come on old friend, let's remake our previous magic.

There's something nice about a fleecy throw, but they are not as cozy as the quilts my wife made for my daughter, but defining that coziness will remain out of reach as far as my words go.

No one can be correct, here. It's just what appeals more or less to you.

Guitars are manufactured from organic material. Even coming off of a factory line, no two guitars are identical because no two pieces of wood are identical.

Even without organic material, no two would be alike due to manufacturing tolerances. Good quality control keeps those tolerances low enough that the differences are not significant or noticeable.

Any guitarist that places outside of the house will have no problem inflicting enough damage on their guitar to make it unique.

Wouldn't a better example than guitars be violins? A handmade violin is always going to sound better than one machinemade because the luthier is able to treat the wood specifcally to showcase the tone of the wood used.

Do I want to take medication produced by the same process that leads to unique sounding guitars?

> If that claim about hand made items being better were true, there would be a market for hand made CPUs

Isn't a CPU kind of one of the most complex thing humanity has ever created in hardware terms? Like the scales involved are so different, compared to the examples of a guitar, chisel or knife. It requires so tiny tolerances and measurements, that I guess we're past it's even possible to "make a CPU by hand" that comes close to our current CPUs. Sure, you can always use old CPU designs from back in the day, but a knife today can be the same as it was 2000 years ago, it just has to be sharp and comfortable, so really hard to compare.

> I have seen this sentiment expressed, but I have yet to see any real evidence for it.

Have you gone searching for evidence? The examples being guitars, chisels, knifes, suit and shoes, sounds simple to search how the top-of-the-line stuff is made. I'd bet on that high quality things from those categories are in fact typically hand-made.

Nobody expressing the general sentiment of “hand made is intrinsically better than factory made” has been able to produce evidence of intrinsic superiority. The burden of proof is on those making such claims and evidence largely favors factories as producing superior products. There can be exceptions, but that is due to bad decisions rather than some intrinsic issue with the idea of a factory production line.

> I really don’t expect to learn any differently by doing further research, which I expect to be a waste of my time.

I guess that settles that, then.

I had moved the shoe anecdote before I saw your reply since it fit better elsewhere and I did not want to write the same thing twice:

https://news.ycombinator.com/item?id=42544305

Anyway, why waste my time doing further research just because a random guy who convinced himself arguably inferior hand made goods are superior to factory made goods did not like hearing that? The opportunity cost is not worthwhile:

https://en.wikipedia.org/wiki/Opportunity_cost

Finally, “if you have not done enough research to agree with me, you have not done enough research” is a fallacious line of thought, since it excludes the possibility you are wrong. Coincidentally, I have an interest in this particular line of research since it is hard for me to find shoes that fit. That is why I have several pairs of the same exact shoe in my closet in the first place. In any case, I have done enough research to be comfortable. The fact that the results do not align with your beliefs is your problem.

> I would rather have factory made versions since I know they are likely made precisely to a specification and deviations from that specification likely make things worse, rather than better.

For these items in particular, there is generally no extant factory automation that will get you a better result than a human craftsman. You'd be a fool to pick extant mass produced alternatives, all else being equal.

A CPU is a pretty poor analogy here, since it's not only precision that makes handmade examples of the above items superior, but it's essentially the only thing that matters for a CPU (or something like an iPhone, etc). You likely could design a superior factory for many of those items, taking into account materials variance, bespoke details for individual customers, etc., but it would not be economical in the markets where these items are sold.

edit: I also think these items are a bad analogy for medicine, where precision is paramount. So I probably agree with you there.

There are plenty of factory made shoes that have fancy contoured uppers that have shapes that can’t really be achieved by hand.

> I have seen this sentiment expressed, but I have yet to see any real evidence for it.

This boils down to the definition of what "better" means, and for items like those listed that's actually quite easy! "Better" means "the way I want it". Customized to my individual needs. Factory produced can never give you that.

If I want a Gyuto style knife with an octagonal handle made from carbon fibre and blade in CPM-10V alloy, I'm not going to find it at Ikea.

If I want a suit or a leather boot that fits my particular body perfectly, it needs to be custom made.

If I want a built K20 turbo engine, it will need to be very different if it goes in a circuit racing car or a time attack car, and both need to be very different from the factory engine.

The factory needs to optimize for the average person. It needs to satisfy regulators, beancounters, shareholders. The artisan does not.

You are directionally correct but the top of the line chisels are in fact hand made in Japan and I suspect the same for knives. Lee Valley and some other higher end manufacturers make some damn fine chisels but chisels are hand tools and I would guess 9/10 woodworkers who use chisels will choose hand made Japanese chisels over any factory manufactured tools.

Isn’t the main thing that makes a chisel better or worse the material quality, such that regardless of whether it is hand made or factory made, the chisel made of superior material is better?

Selection of metal, lamination, presence of a ura (裏), handle material, the ability to make a new handle and attach it to the original metal. A metal chosen for an extreme sharpness may be terribly difficult to sharpen, a metal easy to sharpen may not hold an edge for long.

Most factory chisels are not laminations like hand made Japanese ones are. They are just a single alloy.

> a pair of shoes

>> For all of the things listed, I would rather have factory made versions.

I think people believe this who haven't ever experienced truly custom hand-made things. The process of creating an actual "custom, hand-made, artisan-crafted, boutique" shoe will involve first creating a custom "last" - a wooden mold to build the shoe around.

This last isn't literally a mold of your foot, it is a custom-sculpted shape that is specific to both your foot and the particular shoe style. The shoemaker will keep all of their customer's lasts in a large library, so once your last is customized properly for you (in-person), you don't have to return to order more of that type of shoe - you can place future orders in different colors/materials and the shoe will be created around that same wooden last, ensuring a very specific fit that is most comfortable for you.

A factory cannot do this.

The point of hand-made things is customization that goes far beyond what a factory could ever do. If you're buying hand-made things that aren't truly customized for you, then you're largely missing the point.

I recall my father once took my discarded factory made Rockport shoes out of the trash to have the soles repaired by a cobbler in China. From what my father told me, the cobbler claimed he had never seen such high quality shoes. My father thought he was doing me a favor. Unfortunately, I had several pairs of identical shoes in my closet prepared in advance for when the soles wore out.

If he had never seen such high quality shoes, I certainly do not expect him to be able to produce equal quality shoes even if he is able to make a better fit. However, they make enough sizes of shoes that you can approximate the right fit and actually getting the right fit is not significantly different. I have wide feet, which makes getting shoes that fit well difficult since the wide variety is not often shipped to stores and likely has limited production. That is why I had so many duplicates of the same shoe in my closet in the first place. Having found shoes that fit, I do not feel any need for a custom fit.

That said, a custom fit is a very different thing than a claim of being intrinsically better. If we go by what my father claimed the cobbler in China said, the factory made shoes are the better ones (or at least the Rockport ones I wear are).