Exactly. I'm glad we helped. Development, also called rollout, makes more sense, but we just don't do that in the US. Either development or gear inches are good measures because they are independent of wheel size. I've used gear inches all my life, so in my mind, there is a scale of 25 to 130. A 25" gear means I can climb any slope while carrying cargo. A 130" gear is useful if I'm going very fast down a steep, long hill and I want to add even a little more speed. Most people are comfortable cruising on flat ground in a 70" gear. If you have a single speed bike and live in a flat area, you'll probably want your gear to have 70".

It might help to convert these numbers into development distances if you want to think of it that way, but to compare your gearing to someone else's, you may have to speak their language. You might ask a friend, "What gear did you use to climb that hill?" They might say 40 inches. Does that make sense to you? They might instead say 42/28, and if you both have 700c wheels it makes sense, as it's a 40" gear. But I don't like the tooth count measuring method because it relies on the wheel size for it to be uniform. Besides, as I pointed out, it's hard enough to calculate gear ratios in your head, it's even harder to remember which ratios are equivalent. 42/28 is nearly the same as 52/34, but who remembers that? So when someone says he climbed a hill in his 42/24 gear, I ask what that is in gear inches.

I'm glad I've switched completely to IGHs.

My Rohloff14 is awesome for hills and SA RD5w is awesome for thrills. It has my 5 longest and fastest records.
Nothing gets me down the highway easier. ZERO worries now.

Can't say I have any experience with modern Sturmey Archers, and I haven't ridden any of them at all since I was about 8 a long time ago (and it was a three-speed). I have now-an-older Bullitt with a Shimano 11-speed, and my most recent one - an Omnium Cargo with Rohloff. Both are brilliant, but I much prefer the Rohloff - not least because that one also has a belt-drive (the old Bullitt frame did/does not accept belts- newer models does).

Gear inches is the best. That's all I use.

And that is exactly what I'm saying. This calculator by default is for a 6sp 53/42 setup. And it talks about freewheels, not cassettes. While the calculator works up to 10 or was it 12 speed cassettes, what is the point? There would be more usable gears using 1 chainring and the 10, 11 or 12 cogs you have in the rear and never go into another chainring. And it just occurred to me that is why 1x11 is getting popular.

I'm not sure what your beef with that calculator is? There are losses (and wear and tear) if you use the big chanring with the big sprocket and vice versa. Hence the "shift pattern" which will help you avoid most of those.

(Edit: That said, you could go with an IGH or a 1x setup if you prefer derailleurs).

As a singleseeed guy, I've found that gear inches is a common measurement language among other riders. It's very useful when pursuing the right combination of chainring and cog.

Not true. You just explained what the Fahrenheit scale is based on, so clearly not “nothing”. Fahrenheit’s zero was the coldest he could achieve using available materials and techniques at the time and is a good choice: other workers would be able to achieve the same temperature, using his method, and this would aid communication among scientists working in different parts of the world. It was a good start.

But let’s not get distracted by details of early development which seem quaint to us today. The importance of having an instrument that could put a numerical quantity to the quality of “hot” and “not so hot” was “paradigm-shifting”, as the biz-book writers today call every management fad that comes along. It became possible to realize that heat and temperature were not the same thing. This led to classical thermodynamics and later to modern statistical mechanics. Recognizing the value of a thermometer, and inventing one that would work, was ground-breaking.

The “metric” system is based on the metre, but only for length. The original definition of the metre is no less arbitrary than the length of King Henry II’s nose to fingertip to define the yard. The important thing is that everyone agree on the length of a yard or a metre, whether by royal edict or revolutionary terror. The decimal nature of metric length is a convenience for school children who no longer have to memorize how many inches in a furlong. But note that only in length (and area-volume) do we use powers of 10. For time we still use seconds, minutes, hours, etc. in all but the most rigorous scientific work. And for temperature we don’t use powers of 10, either. So for both time and temperature there is no fundamental reason to use one scale over another. The freezing point of water is every bit as arbitrary as Fahrenheit’s NH4Cl brine. And temperature today is calibrated with the triple point of water, not the ice point, because the latter varies with pressure.

So there is no inherent reason to prefer Celsius degrees over Fahrenheit, particularly for ordinary use. As Tom says, 0 to 100 degrees covers the ordinary temperature range encountered by humans in the productive regions of the world. Outside that range it is a challenge to survive, much less thrive and generate tradeable wealth.

When I had 7,8,&10 speed systems, actually never had a 9sp, I did shift the chainring back and forth to match the next gear inch in line though less so with 10sp but only after having integrated levers. Down tube and stem mounted weren't worth the effort. Now with 12sp I still prefer 2x but for a different reason. 1 chainring covers one range of riding and the other covers a different range. Maybe I'll need to shift the crank for certain random conditions but on the whole its nice to have basically 2 ranges of gears for where I need them but it does do away with the constant crank shifting. MTB is 1x since its only used in the woods.

It might be goofy in how it was created but from an experiential perspective it is far superior to celsius. Moving through life 100 is hot, 104 sounds scary and is, 32 is cold, 0 is damn cold and anything negative is just negative to the idea of going out or even waking up for the day. I spent plenty of summers in the Mediterranean and could never grasp the whole celsius range. 37 just doesn't sound hot and the 7 degree range you hear about doesn't convey the true temp difference the way fahrenheit does. When they're saying the temp is going to drop to 65 from 102 during the day you know you need an extra blanket nearby for when you wake up cold from the easy to read 35+ degree temp change. Going from a summer of 97-100 for several weeks straight to a winter of arctic inversions that dropped the temp to -30, the numbers feel more real the reality of those extremes.

Okay, it’s is based on “something” but it is arbitrary. The Celsius system is based on far easier conditions to replicate and far less arbitrary. An ice/ammonium chloride/water system is the only point on the Fahrenheit scale that is fixed. Human temperature fluctuates...Fahrenheit use 96°F for the human body which is somewhat low.

The Celsius scale and the Fahrenheit scale were developed about the same time (about 20 years between them). Claiming that one is more responsible for the development of thermodynamics over the other is a stretch.

No. The metric system is based on length, mass, temperature, time, electric current, luminous intensity, and amount of substance. Each one is based on a physical measurement that can be reproduced in nature. While the length of a meter was originally 1 millionth of the distance on a great circle route from the North Pole to the equator, it is distance that can be measured and doesn’t change. It’s a bit hard to go check to see if the yard really is the distance from King Henry II’s nose to finger tips since he is long dead.

We do use powers of 10 for mass. And we certainly use powers of 10 for derived units.

Why is the freezing point (or triple point to be more precise) of pure water “arbitrary”. It won’t vary and is reproducible with a minimum of effort. The freezing point of a brine solution is going to vary depending on the concentration of the brine.

^ I should have known better and just held my tongue. As usual, you are being argumentative, picking out of context and misinterpreting the inevitable elisions that one has to make for a quick post, just so you can find fault with things not relevant to the Fahrenheit-Celsius discussion.

Once there is an accepted conversion between the two scales, any temperature can be expressed with equal precision in either of them, without loss of computational convenience. Nowadays the metre and second are defined, unambiguously for all time, in terms of subatomic processes. (Mass has proven to be much harder to express this way.) The yard could be similarly defined if we wanted to, but silly to duplicate the effort. The beauty of the metre, and the quantities originally defined in terms of it, (litres and kilograms) was the powers of 10 used to define subdivisions and multiples. But we didn’t go that route with time. We still think in terms of hours and days, nano-seconds but not kilo-seconds. With temperature, degrees are defined as an arbitrary subdivision of an interval between two arbitrary but reproducible (with today’s methods) anchor points. The fact that the subdivision in Celsius is 100 carries no scientific or computational benefit to cause us to prefer it over Fahrenheit. In science we do, just because other scientists do. But we could equally well use Fahrenheit. Converting between the scales is easy and causes no loss of precision because, unlike converting between yards and metres, the defined formula gives an exact answer. Saturated brine with ice gets you a temp cold enough to make ice cream — that’s the take-home.

My point was just that there is no inherent reason of Nature to prefer Celsius over Fahrenheit. Doing so isn’t necessary to get rid of such nuisances as finding three-fifths of 1 17/64ths inches, or drams in hogsheads and grains in ounces (and is that Troy or avoirdupois ounces?). Temperature doesn’t have any of that. All that matters is that the thermometers are accurate in whatever scale they read in. Everything else is just convention for the sake of easy communication.

(A truly rational temperature scale would use as its zero the lowest possible temperature there is — what we now know to be the unreachable Absolute Zero. Then you’d make the degrees any convenient size where one degree change is meaningful— they’d probably be Fahrenheit degrees, not Celsius, too big. So room temp would be about 530 degrees and a really cold day in North Dakota would be 450. Not very convenient for everyday use on weather reports and no scientific advantage, just the cold Revolutionary rationality for its own sake).

No beef with the calculator per se, just that with 10sp and up at least it has become somewhat of a relic. And the big/big combination wear argument has been going on for a while now, and I've come to the conclusion myself that it is just fine to use it. But my point is that the double shift pattern is unnecessary when you have what has become a combination of a "corn cob" cassette with a group of larger cogs added on. I've been riding 13-14-15-16-17-18-19-21-23-26 for years, and recently switched to 13-14-15-16-17-19-21-23-26-29 (these are Campy standard sized 10sp cassettes). I don't think any GI's in between these when combined with a 50 tooth front would make any difference in how I would ride so I never shift onto the small ring for them. I only use the small ring when I want a smaller gear than 50/29, or at least when I'm approaching a hill where I may. I did a quick 20 mile ride yesterday and never went on the small ring, in fact I almost never do. NYC just doesn't have any hills that need it. Only if I cross the bridge to NJ or go upstate is it ever necessary.

I know in the old days people would obsess over it, I'm old myself. When I had a bike custom built for me after I got my first real job in 1979, the shop guy went through all the combinations with me to decide. I still have that bike and the wheels, but I put a 10sp rear wheel and chain on it about 15 years ago and it was really nice to have lots of gears without needing the small ring. It was still downtube shifters but I didn't go down in the front very often. I then bought myself a carbon bike with Chorus 10 that I still use. I also once rented an 11sp 105 setup Specialized Roubaix in San Francisco. That was outfitted with a 11-34 with a 50/34. That 34/34 combo came in handy in SF with the hills. I rode up into Marin County and did some climbs, knowing I had a bail out gear. But I'll say that it would be total overkill with those gears in NYC. I just looked at it on the website.

Just put a little more power into your stroke if you find yourself with a slightly big gear, that's much easier than doing a double shift. Just MHO.

Since we’ve already dragged this discussion far off the topic, I’ll only say that having a temperature scale that is based on easily reproducible physical measurements and is at least partially decimalized trumps one that is based on somewhat arbitrary standards.

I was down my garage today, but I forgot my reading glasses. As an engineer I use metric type scales automatically, as it reduces the number of compensations required in calculations based on pounds and inches. However, while I couldn't see the centimetres on my measuring tape, I could see the inches. :)

I suspect you're speaking as a chemist. Most of the rest of us don't need such precision.

Hehe, I guess the world outside of the US are all chemists.
What's funny is that Americans already know the metric system and everyone uses it: Money. 100 cents to a Dollar.
But when it comes to everything else, somehow there is some hard feelings that makes the switch to metric all around very difficult.

I'm sticking with absolute zero, the triple point of water, and the freezing point of tin. ;)

I tried to set Mike Sherman's to gear centimeters, but it won't do it.

I'm completely metric. That said, I find gear inches useful to compare various gearing.

In my case, with old leg injuries, I have a very limited amount of power I can put down with one leg. Which means I've got a certain range of "low" gearing that'll work for me. Much higher, anything anywhere requiring more force than I've got, then I'm walking.

So, it has that utility.

In my case, anything sub-15 gear-inches allows me to handle ~10-15% grades, if only for relatively shorter distances. Won't ever be a longer "hill climb" for me, anymore, given the issues. Am putting together a bike, right now, that'll end up sub-14 on the "low." Which isn't quite as slow as walking speeds, but it's close. That'll do. (Was hoping for sub-13, but it's reliant on new drive train parts that won't be coming until 2021 sometime, or later.)

On the "high" end, about the tallest "high" I seem to do okay with is: ~80 gear-inches, or so. Anything more, and I'm having to push harder than I can do for more than a few moments, on some terrain. Anything less, and I'm spinning out, on the flats.

So, for me at least, a bike with ~ 14-80 gear-inches works well. It'd drive most people nuts, I'm sure. But it works for me.

Greatest utility: is it low enough; and when comparing one drive train combination to another, to determine "which is better."

It covers any gearing you want. the cogs are the same weather freewheel or freehub.

But nobody talks about freewheels anymore unless you are on the C&V forum. It's about the same as a program telling you run it from the DOS prompt.

This a really good post, IMO.