# New Metric Prefixes Get Bigger And Smaller

It always fascinates us that every single thing that is made had to be designed by someone. Even something as simple as a bag and box that holds cereal. Someone had to work out the dimensions, the materials, the printing on it, and assign it a UPC code. Those people aren’t always engineers, but someone has to think it out no matter how mundane it is before it can be made. But what about the terms we use to express things? Someone has to work those out, too. In the case of metric prefixes like kilo, mega, and pico, it is apparently the General Conference on Weights and Measures that recently had its 27th session. As a result of that, we have four more metric prefixes to learn: ronna, quetta, ronto, and quecto.

Apparently, the new prefixes are to accommodate “big data” which is rapidly producing more data than there are atoms in the Universe. There were actually proposed earlier in a slightly different form but accepted at the conference. Apparently quecca is too close to a Portuguese swear word. So what do these actually mean? A QB (quettabyte) would be 1030 bytes while an RB (ronnabyte) is only 1027.  So 1 QB would be 1,000,000 yottabytes (YB) the previous top of the scale.

On the other end of the scale, the previous tiny measure was the yocto. So a yW or yactowatt, would be 10-24 watts. A rw (rontowatt) is 10-27 and the qw or quectowatt is 10-30. The larger end of the scale seems more practical as now the sun weighs 2,000 quettagrams instead of 2,000,0000 yottagrams. In the old days, we used to mix prefixes like micro-micro farad instead of picofarad, so you could also say it was 2 mega-yottagrams, but that use has fallen out of favor in modern times.

We were sorry the earlier proposal by a student for ronna failed. The proposal would have had the prefix as “hella” as in a “hellawatt.” At least you’d remember it. Apparently, though, it wasn’t a serious contender because the new entries have to have a unique letter and they try to avoid things that measure common physical units although some of the older ones do. So hella, for example, clashes with both henry — the unit of inductance — and hecto, which is an existing prefix. If you write HW that’s a hectowatt, not a hellawatt. At least in ended in “a” and not “o” which indicates a positive power of 10 instead of a negative power, but that wasn’t enough to earn it a spot on the list.

Of course, the real politics in metrology isn’t the naming of things, but rather the definition of the units. Some people really get into that.

## 72 thoughts on “New Metric Prefixes Get Bigger And Smaller”

1. Doug says:

Who the hell is using prefixes after your exponent gets to like 10^+/-18? Other than “science enthusiasts”, of course.

1. paulvdh says:

10+ Terabyte HDD’s are quite common and the first Petabox is now around 10 years old.

In engineering we’ve switched from N/mm^2 to MPa and when you get into hydraulics or Youngs Modulus the numbers get bigger quite fast.

1. HaHa says:

Speak for yourself, PSI was good enough for my grandfather, it’s good enough for me.

1. Gravis says:

Sure…. but isn’t he the one that died because he improperly converted kilometers to miles which made his PSI calculations off which resulted in his submarine imploding?

1. Mystick says:

If it were properly miles to begin with, that never would have happened. The metric system killed that man.

2. Pat says:

Same reason you use any prefix: when you have a characteristic term near it.

Zetta/yotta were both added recently, and exascale computing is close enough that future planning is likely to start using them. So it’s not nuts to plan past it again.

Gets used in particle astrophysics, too. It’s a bit weird to use ZeV for energies when that literally gets you to joule scale, but zevatron has callbacks to the tevatron, so ZeV it is.

2. The Commenter Formerly Known As Ren says:

IIRC, Richard Feynman said something to the effect that we used to call large numbers “astronomical”. After seeing what the USA government debt was at the time, he proposed calling large numbers “financial”.

3. elwing says:

you do realize that scientists works for papers and publications, having words means a lot for them.

2. TG says:

Should just replace ’em all with _e¹, _e², _e³ etc. Just tell me how many places to move the decimal and in which direction. Nobody wants to remember where yotta or ronto or quetta is on the scale and in what order. That kind of nonsense seems very un-metric in spirit. Making a bunch of arbitrary stuff to memorize instead of just mapping it clearly in base 10.

1. X says:

So remembering arbitrary integers is easier than remembering arbitrary names? We use names precisely because they are easier to memorize than numbers. This is why Dave is Dave and not “individual #723”.

As far as “un-metric in spirit” goes, the metric system has added more prefixes on multiple occasions. We all had to memorize what femto and atto mean, and nobody had any problems with that.

“The spirit” here is being able to talk about the units of measure in actual use, so this is all completely appropriate.

As our knowledge of the universe expands, so must our vocabulary to describe it.

1. Ha30aH says:

Actually, I did exactly that. At IT corporation where I used to work I renamed my coworkes to numbers, back when Skype 7 was still used and it let you rename contacts. It helped me focus on work because I didn’t have to deal with humans but rather… something like NPCs in game. I know it sound horrible but really, when writing embedded C code I needed full focus and not think about some Johns and Janes discussing irrelevant things on some group.

2. miked says:

Exactly

3. Pat says:

Just do what they did before: to add a power of ten, you just say “meter-six” (for 10^6 meters) and for a negative power, “sixth-gram” (for 10^-6 meters).

You occasionally see that back in pre-1900 papers.

4. RunnerPack says:

Following this logic, we should replace the units themselves with “length unit”, “volume unit”, etc. All those names are too hard to memorize.

3. craig says:

I’m using hellawatt anyway.

1. What’s the conversion from hellawatt to jigawatt and googolwatt?

1. HaHa says:

Useful units.

Juggoliter: The amount of beer needed to get an ICP fan good and drunk.
Footer: 1/3 of a meter. pl Feeter. e.g. I’m 6 feeters tall.

1. Alan says:

If we are brave enough to have a programming language called Brainfuck, we should be able to have a small unit called:

the Fuck-all.

We could use this, to measure how much MOST people care about new SI units.

1. ian 42 says:

it could be an arbitrary number from the size of an electron to the USA debt – where you don’t really care what the number is.. The number in front could be how much you don’t care. ie 10Fa means a lot less care worthy than 1Fa..

4. Andrzej says:

“Apparently, the new prefixes are to accommodate “big data” which is rapidly producing more data than there are atoms in the Universe”.

So, where are they storing the data? The ever-mysterious cloud is still made of atoms, right?

1. Pat says:

They aren’t storing it. The data’s ephemeral: an exaflop over a year produces around 30 yottaflops, total, which is starting to pop the top of the scale.

2. Ray' says:

With 10^80 atoms, you would only need log2(10^80)=266 bits to count all of them, so depending on the context, running out of nomenclature isn’t hard sadly

3. Comedicles says:

We store the data in the optical equivalent of a tape loop. Transmit the data to a reflector a couple light-years away. But the loops are full of energy and we are going to run out of that.

4. cory says:

I came here looking for this comment because of the stupidity of the comparison. This should be obvious to the HaD reader, but the comment above about pop the top of the scale replied to a comment about the lunacy of comparing to the number of atoms in the universe says otherwise (perhaps the commenter meant to reply to the comment about why we need bigger prefixes? – I hope so!)

But, just in case, let us say we generated a quectobit of data/day since the 1960’s. Say 60 years for a nice even divisible by 10 because I am a human, and we’re already working in base 10 with the prefixes. that’s 21900 days (not including leap days/seconds/etc – remember, ease of calculation, it will be on the same magnitude I promise!) that’s on the magnitude of 10^4. So, 10^30*10^4 is 10^34. Thats nowhere near 10^80. Let us say we were producing a quectobit of data per day since ancient egypt. 3100bc. that’s 1,869,530 days, so on the order of 10^6, or 10^36 total bits. Still nowhere near 10^80. Now, let us go to the extreme. Say we’re producing a quectobit of data per second from ancient Egypt. Stay with me, that’s 161,527,392,000 seconds. Looks big right? that’s on the order of 10^11. So, yup, that’s 10^41. And, again, this should be obvious, that’s not better than half-way to 10^80 on a linear scale. Don’t mental shortcut and say we’re only 10^39 bits away from that if we produced that much data per second since ancient Egypt. We add exponents when we *multiply*! The difference is so vast, I’m not even certain astronomical measures would cut it without measuring the size of the universe in femtometers….(hmmm 92 billion light years in diameter, that’s 10^11 light years, 9,460,800,000,000,000 meters, 10^16 meters rounding, putting that in femto…is adding 15 to the prefix, so, 10^31, so there the universe is roughly 10 quecto femtometers in diameter, so, no. Astronomical comparisons won’t work either, nowhere near 10^80. Its even lunacy to try that comparison considering atoms are picometers in diameter, and there are 1000 femtometers in a picometer, and hydrogen is very roughly (rounded up to) 10 picometers in diameter) Hmmm….maybe this comparison would work, if we had 1,000,000,000 quectocomputers (10^9 * 10^30 computers) capable of producing a quectobit of data/second running since ancient Egypt (10^9*10^30*10^30*10^11 = 10^80 ) we would have approached (well, without the rounding exceeded) 10^80 bits calculated.

Now, to really drive the point home…if we managed to make these quectobit/second computers out of a mere 1g of silicon, we’d need 10^39 g of silicon to make the computers needed for this ludicrous scenario. To put it in more palatable units, lets divide by 10^3 to get kilograms. 10^36Kg of silicon. A metric ton lets us drop the exponent by 3 again, so we can say 10^33 metric tons of silicon. According to wikipedia, we refine roughly 1.5 metric tons of silicon per year to metallurgical grade, and 15% of that to semiconductor grade. For fun, we’ll say we do 1.5 metric tons of silicon to semiconductor grade per year, and again just stick to the order of, so 10^6 tons/year. At that rate, we would have needed to be refining silicon for 10^27 years prior to ancient Egypt to make our 1,000,000,000 quectocomputers. since the age of the universe is on the order of 10^10 years, we’d have needed the big bang to create our computers for us to reach that target or have massively increased the amount of silicon we could refine. However, the earth’s crust is roughly 28% silicon, so at an estimated mass of the crust on the order of 10^20Kg or 10^17 tons (so 28% of that, we’ll leave it on the same order). So even stripping the entire crust of the planet would not satisfy the need for silicon for this ludicrous number of computers. So, we’d need the big bang to not only have created our computer for us, but to have provided a small moon of a pure silicon for us to have enough. (Note, I used 1 g as an order of for my order of calculations as the best estimate I found online said 5g is currently what most of our chips use).

Hopefully this drives the point home that we’re nowhere near the estimated number of atoms in the universe. The comparison must have been made thanks to a mental shortcut, ignorance, or plain hubris. (or all of the above). (as even hyperbole doesn’t come close) Regardless, it’s so flawed, it’s hard to describe just how flawed it really is in any meaningful way. Everything I try leaves us with unnamed 10^x values, where people have a hard time understanding 10^9, let alone government-like numbers like 10^15. (1 trillion – as in Teradollars) I guess, the one thing I can do is demonstrate the difference in small numbers. Say, the difference between 10^3 and 10^9. It’s not 1/3 of the way there. 10^3 in 1000, where 10^9 is 1,000,000,000, for a difference of 999,999,000. So it should be obvious that 10^30 is not a significant fraction of 10^80 by any stretch of the word significant. Even if you increased the number of calculations by a million, 10^9, you’re still nowhere near. Even if you hit 10^79, that’s not 1000 away from 10^80, just like how 10^8 – 100,000,000 is not 10 away from 10^9 1,000,000,000, its 900,000,000 away. So even 10^79 is not “really close” to 10^80.

TLDR; We’re nowhere near advanced enough to be comparing anything we do to 10^80. Anything. No excuse for the comparison either, as hyperbole would need an unnamed prefix to describe how hyperbolic it is!

1. RW ver 0.0.3 says:

Well yes, but my brain went “What? It takes several atoms still to store a bit, only a tiny percentage of the atoms on this planet are employed in data storage, therefore must be wrong by dozens of orders of magnitude.”

5. Carl Vehse says:

“Ronna” and “Quetta” – These are now the congressional goals for use as prefixes to future federal budgets. The prefixes, ronto and quecto, will be used when comparing the current buying power of your dollar to that from the previous decade.

1. Dave says:

But because we can’t use the SI system in the US, they have to be multiplied by 872.9 and purple, and divided by 393/7 and llama, respectively.

1. I cringe when I hear actual Americans (in this case: This old Tony) use fractions of an inch and “thou” *in the same measurement*. I can’t find the exact vid but it was something like “three five sixteenth and 6 thou” (3 5/16″ + 0.006″ (just say 84.29mm)

1. X says:

What is the next bigger socket after 3/8″??? Think quick! Crunch that fraction.

What is the next bigger socket after 11mm? So easy!

1. Y says:

Answer: whichever the next smallest bit in the box is.

2. Duh. 4/8″.

I grew up with rulers that are marked in 1/8ths and 1/16ths, so it’s “natural” to me. If I grew up metric, I can see how it would be a head-scratcher.

The secret benefit of calling it out in fractions is that you also get a feel for how precise it needs to be, without resorting to all those zeroes. If you ask for 7/64″, I know you mean business, because anyone sane would have just said 1/8″ if they didn’t.

Also, powers of two are kinda nice. Quick, what’s 0.1 in binary!

3. Joseph Eoff says:

7/16 inch is next up from 3/8 inch in a typical SAE socket set.

12mm is next up from a 11mm in a typical metric socket set – except that is so rarely used I don’t know why they bother to make bolts, nuts, or sockets in that size at all.

You’ll more often see 10 mm and 13 mm. 11 mm and 12 mm not so much.

2. smellsofbikes says:

I’m looking at this as a dare, and am all ready to make a video where I say something’s diameter is 20mm plus or minus a thou. That’ll show ’em.
(Horrible reality: Italian race bicycles up into the 2010 era at least have the bottom bracket standard as 36mm x 24tpi. It’s the only thing I know of where a single feature is both metric and imperial.)

1. HaHa says:

Almost all metric bolts are in even tpi.
As a matter of pride, they don’t express it that way anymore. But do the math.

They’re stuck with it for the foreseeable future. Neener SI dweebs!

2. “20 mm plus 3/8 of a thou” please!

3. Skyler says:

I was doing some work on an old Ford F150 and the exhaust system was held on by 1/4″-24 bolts with 10mm heads. Not quite the same feature, but at least the same part.

2. X says:

Old whine in a new bottle.

“big data” which is rapidly producing more data than there are atoms in the Universe [citation needed]

1. Carl Vehse says:

After that a question might be when will the cardinality of the “big data” set go from countably infinite (Aleph-null) to uncountably infinite (Aleph-one or larger).

1. mathman says:

LoL, makes for an interesting thought experiment!

2. robertrapplean says:

That was a notable exaggeration. There’s something around 10^80 atoms in the known, observable universe, but I also had to look it up.

1. smellsofbikes says:

I was kinda wondering how you store or refer to more data than the atoms available for representing that data.

1. The Commenter Formerly Known As Ren says:

Quarks; Up, Down, Charmed, etc…
j/k

2. HaHa says:

Photons in flight.

3. Al Williams says:
7. Doug says:

Some argue the imperial measurement system is arbitrary, based on the king’s foot or some such thing. I would argue the metric system is also arbitrary. For example, the definition of a meter is 1 10,000,000th of the distance from the North Pole to the equator. The main attribute of the metric (SI) system, is that it operates in multiples of 10. I once had some fun debating a German engineer who was bothered by the USA inability to fully adopt the system. I responded by asking, “Is this because everyone else is less skilled at working with numbers which are not multiples of counting on ten fingers?” We do have other numbering systems available, binary, octal, and hexadecimal to name a few.

1. jbx says:

Don’t USA engineers use bananas as measurement units ? ;)

1. Kaz says:

Yeah, as rulers.

2. ian 42 says:

binary, octal, and hexadecimal are ALL more sensible that what the USA uses…

3. bung says:

I would argue the main attribute of the metric system is that it is not arbitrary. For example 1000kg or water = 1000 liter of water = 1m^3 of Water that freezes at 0C and boils at 100C (at 1 atmosphere)

There is loads more and regularly the maths just makes sense as all the units are related.

1. Pat says:

Almost right.

When the metric system was developed, the point was to democratize units. You’re spot on regarding using water as a mass reference. That was super smart.

But the distance measure (1/10E6 a quadrant) was idiotic. That’s not something people could key off of.

You know how I know? Because they effed up the trig when they did it first. So the meter’s *just slightly* off. But by then they had this great meter reference bar that they had made, and were like “close enough.”

Even stupider, there new reference was like a tiny bit off from a yard anyway. So it was like “change all your standards to… our new pointlessly different standard.”

Grams were a cool idea. Meters were dumb as bricks.

1. Kat says:

It’s a bit depressing that an article about the metric system has a comment section full of people basically gloating about being on the big brain team and trying to harass people into thinking they’re right

That’s how France ended up with ten hours days after all…

2. Comedicles says:

Super smart would have been choosing a standard yard and calling it a meter.

8. Mystick says:

I find it amusing that the hyper-proponents of the metric system end up using it incorrectly so frequently. How far is it to the moon: “Approximately 300000 kilometers”. WRONG. The correct answer is approximately 300 megameters. Use the system or don’t, but do it right.

1. Kat says:

Or even better, .3 gigameters!

2. Chris says:

It’s 300 Mm only if you are an engineer, scientifically it’s 0.3 Gm — gigameters.

3. Conor Stewart says:

That is a little detail that really doesn’t matter all that matters is you are consistent in what what you use, if you use 300,000 Km in a calculation just make sure your output units are kilometers too, doesn’t matter what prefix you actually use, it can be converted to other prefixes before or after the calculation. So I take it according to you, the speed of light being 3×10^8 m/s is actually wrong and it should be 300 Mm/s?

9. Hassi says:

they had the chance and threw it away!

ligma

P2: “I need a light bulb”
P1: “Certainly, how many Watts?”
P2: “ligma”
P1: “ligmaWatt”?
P2: “exactly”

10. Aww, no :(. Why is it not Zetta-Yotta-Quetta-Ronna? That’d sound much nicer.
Now it’s much more ‘artsy’, like a haiku or something.
I’m doomed to mix them up for all eternity…

11. shod says:

You should not call ISO ‘metric’
Metric means based on the meter, and of course many standard values are, but as other commenters pointed out when amusingly using these new prefixes for the US budget, these are just a number definition which can be used for anything including non-meter derived standards.
And the US also uses ISO as standard of course and in fact they are ISO members and have voting rights in ISO.

12. BaZ says:

A QB (quettabyte) would be 1030 bytes while an RB (ronnabyte) is only 1027. So 1 QB would be 1,000,000 yottabytes (YB) the previous top of the scale.

is one tag that shouldn’t be stripped…

13. Comedicles says:

The largest and smallest sounding similar, and differing by a typo, would make a much more dramatic “Spinal Tap” effect.

1. Mystick says:

“Our large megalithic monument was in danger of being crushed by a dwarf!”

14. Dean Drover says:

Question: on a good, metal ruler, is that black ink in the grooves or just shadow? I bought an oldie and a goodie from a tool thrift Restore and I really can’t seem to tell without buying a microscope. I get this is off topic, but saw the ruler picture…and so, why not ask the pros.

1. None says:

Use a flashlight and hold it close and at an angle, that should show if it’s just a shadow or ink (or well dirt…).

1. None says:

If you mean the article picture, the groves are filled with ink. You can tell that the ruler lines that seem lighter are so due to reflection of light, similar to how it’s noticeable for the numbers, which are much larger in size, and which are black except for the corners/curves where they are shiny (reflection).

15. Mike says:

If data takes at the very minimum of 1 atom per bit to store, then all the data in the world would consume equal amounts of atoms, then there are the rest of the atoms in the universe, so thusly, data can not exceed the atoms of the universe as data is within the universe

1. Mystick says:

It’s the “dark energy”. We have styrofoam packing peanuts being the “missing matter”. Some ancient alien civilization had a galactic-level YouTube server and that’s what started the whole “expansion phase” of the Universe seen in the current era.

Physics == Solved

16. MartyK says:

New Metric Math
~~~~~~~~~~~~~~~
– Ratio of an igloo’s circumference to its diameter: Eskimo Pi
– 1000 kilos of chinese soup: Won ton
– Time between slipping on a bannana peel and hitting the pavement: 1 bananosecond
– Weight an evangelist carries with God: 1 billigram
– Half of a large intestine: 1 semicolon
– 1000 aches: 1 kilohurtz
– 1 million microphones: 1 megaphone
– 2000 mockingbirds: 2 kilomockingbirds
– 52 cards = 1 decacard
– 10^6 bicycles: a megacycle
– 10 rations at Christmas: 1 decoration
– 3 statute miles of intravenous surgical tubing at Yale University Hospital: 1 I.V.League
– basic unit of laryngitis: 1 hoarsepower

1. Onetruegod says:

(slow clap)

17. Manfred says:

“producing more data than there are atoms in the Universe.”

A bit can not be stored in less than an atom.
Therefore there can never be more data than atoms in the universe.

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.