Precision, Rate and Accuracy in watches: a layman's view
- Amor Vincit Omnia
- Moderator
- Posts: 33793
- Joined: Sat Jun 11, 2011 7:34 pm
- CW-watches: 4
- Location: Norfolk, UK
Precision, Rate and Accuracy in watches: a layman's view
I’m no horological expert, so please excuse me if I get some of this wrong, and please step in and correct if I make mistakes!
We occasionally get threads on the forum where people talk about the “accuracy” of their watches, especially with watches performing or not within manufacturer’s stated (or COSC) specs.
There are three words we need to understand here: precision, accuracy and rate.
Let’s begin with the obvious. We would all be delighted if all of our watches performed with total accuracy, in other words +/- 0.0 s/d. A watch running at 28,800 bph will need to beat 691,200 times in 24 hours. Add or subtract 8 from that total and you have a rate of 1 second per day. Expressed as a percentage that’s phenomenally high. Even a watch deviating by 5 s/d is well over 99.99% accurate. So why don’t all watches achieve it? Let’s look at the targets on the image.
The answer is down in great part to precision. Think of this as an expression of how well the watch performs in relation to itself. Position and temperature are two of the biggest factors determining this. There is also something called isochronism, which I don’t really understand, but I read that it’s about maintaining an even torque as the spring winds down. The more consistently a watch movement performs over a range of positions, temperatures and states of torque in the mainspring, in theory the more precise it is. The marks on the target are tightly grouped.
In theory, a precise movement will give you a steady and predictable “rate” (average daily gain or loss), which should mean that it is easier to regulate the movement so that it performs closer to +/- zero.
A watch that gives greater fluctuations over the same range will be harder to regulate because it is more random. The better the quality of build and materials, and the more thorough the testing and adjustments, the more precise the movement...in theory!
Accuracy: the measure of how a watch performs against an external criterion: obviously, the time referenced by a radio, internet or some other type of signal. Closely grouped markings (daily rate) near the centre of the target (objective time measurement) give us a watch that is both precise and accurate. A consummation devoutly to be wished, as Shakespeare put it.
If, however, your watch (once set) had recorded the following over 5 days: +7, +2, -12, -4, +7; then your watch would be showing the correct time, and you could claim that it was perfectly accurate. But it wouldn’t be very precise, because its rate (distribution of marks on the target) was all over the place. The accuracy in this case is down to luck, and you can bet that if you left it in the position that gave -12 for 5 days it would probably lose a minute or so. Or it should. In a similar way an imprecise marksman or bowman might occasionally score a dead-centre bullseye on the target.
We talk a lot about COSC specs. COSC (Contrôle Officiel des Chronomètres Suisses) is a test performed at one point in time on an uncased movement, sometimes months before the watch is cased and shipped. Factor in that it may sit in a jeweller’s store for months thereafter. Of course, watches with COSC certification should perform to a given standard, as they are often superior grades of movement. Generally speaking, the better the movement, the more precise and, if well regulated, accurate the watch should be.
Really top-end watches have things like “Adjusted to 5 positions” etched in gold on the movement bridges. In the case of manufacturers like P-P and V-C this is done in-house and takes a lot of time, along with the superb quality and construction of the movements in the first place. The movements are capable of both precision and accuracy and in general they seem to be tuned accordingly. There is, of course, a cost element involved.
Anyway, to specifics. If you know the rate of a watch (i.e. if it is precise) you can adjust accordingly. My Speedmaster has always gained 8-9 s/d, no matter how I wear it or what position it is left in. It is therefore very precise. It isn’t that accurate. The latter doesn’t bother me too much; I can set it to -30s on a Sunday, confident that by Wednesday it will be within a few seconds of “true time” and by the following Sunday it will be about 30s fast, at which point it will be taken back by a minute. It’s 4 years old and when a service falls due I will ask them to regulate it a bit better.
Interesting that the two most precise mechanical watches I own are my two most expensive. The other, which never gains more than +5s per week, is that well-known piece of overpriced rubbish, the Cartier Tank. Food for thought, eh?
We occasionally get threads on the forum where people talk about the “accuracy” of their watches, especially with watches performing or not within manufacturer’s stated (or COSC) specs.
There are three words we need to understand here: precision, accuracy and rate.
Let’s begin with the obvious. We would all be delighted if all of our watches performed with total accuracy, in other words +/- 0.0 s/d. A watch running at 28,800 bph will need to beat 691,200 times in 24 hours. Add or subtract 8 from that total and you have a rate of 1 second per day. Expressed as a percentage that’s phenomenally high. Even a watch deviating by 5 s/d is well over 99.99% accurate. So why don’t all watches achieve it? Let’s look at the targets on the image.
The answer is down in great part to precision. Think of this as an expression of how well the watch performs in relation to itself. Position and temperature are two of the biggest factors determining this. There is also something called isochronism, which I don’t really understand, but I read that it’s about maintaining an even torque as the spring winds down. The more consistently a watch movement performs over a range of positions, temperatures and states of torque in the mainspring, in theory the more precise it is. The marks on the target are tightly grouped.
In theory, a precise movement will give you a steady and predictable “rate” (average daily gain or loss), which should mean that it is easier to regulate the movement so that it performs closer to +/- zero.
A watch that gives greater fluctuations over the same range will be harder to regulate because it is more random. The better the quality of build and materials, and the more thorough the testing and adjustments, the more precise the movement...in theory!
Accuracy: the measure of how a watch performs against an external criterion: obviously, the time referenced by a radio, internet or some other type of signal. Closely grouped markings (daily rate) near the centre of the target (objective time measurement) give us a watch that is both precise and accurate. A consummation devoutly to be wished, as Shakespeare put it.
If, however, your watch (once set) had recorded the following over 5 days: +7, +2, -12, -4, +7; then your watch would be showing the correct time, and you could claim that it was perfectly accurate. But it wouldn’t be very precise, because its rate (distribution of marks on the target) was all over the place. The accuracy in this case is down to luck, and you can bet that if you left it in the position that gave -12 for 5 days it would probably lose a minute or so. Or it should. In a similar way an imprecise marksman or bowman might occasionally score a dead-centre bullseye on the target.
We talk a lot about COSC specs. COSC (Contrôle Officiel des Chronomètres Suisses) is a test performed at one point in time on an uncased movement, sometimes months before the watch is cased and shipped. Factor in that it may sit in a jeweller’s store for months thereafter. Of course, watches with COSC certification should perform to a given standard, as they are often superior grades of movement. Generally speaking, the better the movement, the more precise and, if well regulated, accurate the watch should be.
Really top-end watches have things like “Adjusted to 5 positions” etched in gold on the movement bridges. In the case of manufacturers like P-P and V-C this is done in-house and takes a lot of time, along with the superb quality and construction of the movements in the first place. The movements are capable of both precision and accuracy and in general they seem to be tuned accordingly. There is, of course, a cost element involved.
Anyway, to specifics. If you know the rate of a watch (i.e. if it is precise) you can adjust accordingly. My Speedmaster has always gained 8-9 s/d, no matter how I wear it or what position it is left in. It is therefore very precise. It isn’t that accurate. The latter doesn’t bother me too much; I can set it to -30s on a Sunday, confident that by Wednesday it will be within a few seconds of “true time” and by the following Sunday it will be about 30s fast, at which point it will be taken back by a minute. It’s 4 years old and when a service falls due I will ask them to regulate it a bit better.
Interesting that the two most precise mechanical watches I own are my two most expensive. The other, which never gains more than +5s per week, is that well-known piece of overpriced rubbish, the Cartier Tank. Food for thought, eh?
Steve
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
-
- Senior Forumgod
- Posts: 1406
- Joined: Thu Jul 16, 2009 9:20 pm
- CW-watches: 6
- Location: Somerset
Re: Precision, Rate and Accuracy in watches: a layman's view
As a fellow layman this seems like a good summary and makes perfect sense throughout - good work
- missF
- CW Forum Poet Laureate
- Posts: 11866
- Joined: Thu Mar 06, 2014 2:59 pm
- CW-watches: 3
- Location: Edinburgh
Re: Precision, Rate and Accuracy in watches: a layman's view
Precisely.
Well written, thanks
Well written, thanks
watching you fail in your quest for a “one watch” has been great entertainment
Watchaholic
‘Imprudently spendy’
Thomcat00
Watchaholic
‘Imprudently spendy’
Thomcat00
Re: Precision, Rate and Accuracy in watches: a layman's view
Great write up. The thing I find most fascinating though is the revelation that you’ve had your Speedmaster for four years!!!
I remember you revealing that on the forum and it can’t really be that long ago can it?
I remember you revealing that on the forum and it can’t really be that long ago can it?
Iain’s Law: Any discussion on the Christopher Ward forum, irrespective of the thread title or subject matter, will eventually lead to someone mentioning the Bel Canto if the thread continues for long enough.
- Amor Vincit Omnia
- Moderator
- Posts: 33793
- Joined: Sat Jun 11, 2011 7:34 pm
- CW-watches: 4
- Location: Norfolk, UK
Re: Precision, Rate and Accuracy in watches: a layman's view
^^^March 2017, Iain!
Steve
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
- copemanphoto
- Senior Guru
- Posts: 593
- Joined: Tue Apr 25, 2017 8:37 pm
- CW-watches: 2
- Location: Fort Saskatchewan , Alberta, Canada
Re: Precision, Rate and Accuracy in watches: a layman's view
Thank you for the great write up.
Cheers, Mike.
- Noush
- Senior Forumgod
- Posts: 1331
- Joined: Sun Jan 19, 2020 12:30 am
- CW-watches: 2
- Location: Blue Mountains Australia
Re: Precision, Rate and Accuracy in watches: a layman's view
Nicely put together Steve and thanks for making the effort!
Chris
"Life is what happens when you are busy making other plans" : John Lennon.
"Life is what happens when you are busy making other plans" : John Lennon.
- ajax87
- Senior Forumgod
- Posts: 3453
- Joined: Sat Aug 01, 2009 2:47 am
- CW-watches: 7
- Location: West Michigan, USA
Re: Precision, Rate and Accuracy in watches: a layman's view
I appreciate this. My mind has danced around this but hasn’t been able to land until reading this. The visual is great!
Alex
C5A Mk1|C65 316L LE|C63 36mm, GMT, Elite, 2023ish FLE|C1 Moonglow|Omega Seamaster DeVille|Speedmaster Racing|MoonSwatch Mercury|RZE Endeavor|Tudor BB58 925
C5A Mk1|C65 316L LE|C63 36mm, GMT, Elite, 2023ish FLE|C1 Moonglow|Omega Seamaster DeVille|Speedmaster Racing|MoonSwatch Mercury|RZE Endeavor|Tudor BB58 925
Re: Precision, Rate and Accuracy in watches: a layman's view
Makes sense of the terms. Well done.
"Nothing in all the world is more dangerous than sincere ignorance or conscientious stupidity."
-Martin Luther King Jr.
-Martin Luther King Jr.
- Bahnstormer_vRS
- Moderator
- Posts: 35161
- Joined: Thu Sep 06, 2012 3:06 pm
- CW-watches: 34
- LE-three: 1
- LE-foura: 1
- LE-fourb: 1
- LE-six: 1
- Location: Hertfordshire, UK
Re: Precision, Rate and Accuracy in watches: a layman's view
An interesting analysis Steve, that makes sense, much as I pay little attention to the concepts of accuracy or precision you portray, as I generally only wear a given watch for a couple of days at most.
That said, I think I would prefer my watches to have 'high precision' as I can more easily correct for accuracy myself.
Guy
Sent from my Xperia 5 using Tapatalk
That said, I think I would prefer my watches to have 'high precision' as I can more easily correct for accuracy myself.
Guy
Sent from my Xperia 5 using Tapatalk
In small proportions, we just beautie see:
And in short measures, life may perfect bee. - Ben Jonson (1572 – 1637)
Inscription on the Longitude Dial
Hatfield House, Hatfield, Hertfordshire AL9 5NB, England
And in short measures, life may perfect bee. - Ben Jonson (1572 – 1637)
Inscription on the Longitude Dial
Hatfield House, Hatfield, Hertfordshire AL9 5NB, England
- Amor Vincit Omnia
- Moderator
- Posts: 33793
- Joined: Sat Jun 11, 2011 7:34 pm
- CW-watches: 4
- Location: Norfolk, UK
Re: Precision, Rate and Accuracy in watches: a layman's view
I used to wear my watches pretty much as you do, Guy, but over the last year or two I have worn my "main" watches for longer periods, often a week or two, without a change. So the concept has become more important to me.Bahnstormer_vRS wrote: ↑Sat Apr 10, 2021 11:56 pm ...much as I pay little attention to the concepts of accuracy or precision you portray, as I generally only wear a given watch for a couple of days at most.
That said, I have slightly fewer watches to rotate than some people.
Steve
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
- H0rati0
- Senior Forumgod
- Posts: 2314
- Joined: Wed Jun 27, 2018 7:49 am
- CW-watches: 4
- Location: Alpenvorland
Re: Precision, Rate and Accuracy in watches: a layman's view
Thanks for pulling this together as a nice summary Steve - a useful reference!
nik
nik
"There is no beginning to enlightenment and no end to training" - Dogen Zenji (1200-1253)
- strapline
- Senior Forumgod
- Posts: 2328
- Joined: Tue May 07, 2019 3:00 pm
- CW-watches: 0
- Location: SW Ireland
Re: Precision, Rate and Accuracy in watches: a layman's view
Thanks for this explanation of precision and accuracy, Steve. You once offered up a way shorter version of this when I enquired as to whether a watch I had was running within acceptable timing tolerances. At this point I hadn't considered the notion of precision, and its relation to a watch's performance, too. To quote your calculations:
Isn't this slightest of deviations of a mechanical movement where their beauty and appeal lies? Imagine if all mechanical watches were bang on to the second...every day, every year. Wouldn't you then end up with something way more sterile than a quartz watch? In short you would end up with perfection, which I think would be a very uncomfortable place. It would certainly change the appeal of watch collecting as we know it.
Des
So, a watch running fast or slow by 5s a day will be out by 40 vibrations out of a daily total of over 691K. That is frankly remarkable; and it repeats that feat day after day, after day.A watch running at 28,800 bph will need to beat 691,200 times in 24 hours. Add or subtract 8 from that total and you have a rate of 1 second per day.
Isn't this slightest of deviations of a mechanical movement where their beauty and appeal lies? Imagine if all mechanical watches were bang on to the second...every day, every year. Wouldn't you then end up with something way more sterile than a quartz watch? In short you would end up with perfection, which I think would be a very uncomfortable place. It would certainly change the appeal of watch collecting as we know it.
Des
Last edited by Amor Vincit Omnia on Sun Apr 11, 2021 10:22 am, edited 1 time in total.
Reason: Sorted out quotes
Reason: Sorted out quotes
Does melancholy count as two of your five daily servings?
- Amor Vincit Omnia
- Moderator
- Posts: 33793
- Joined: Sat Jun 11, 2011 7:34 pm
- CW-watches: 4
- Location: Norfolk, UK
Re: Precision, Rate and Accuracy in watches: a layman's view
On a movement beating at 28,800 bph, yes. The Omega 1861, for example, beats at 21,600 (6 beats/sec as opposed to so the figures are different. Still phenomenal, though.
Steve
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
Linguist; retired teacher; pilgrim; apprentice travel writer
Take nothing but pictures, leave nothing but footprints, kill nothing but time
Avoid loud and aggressive persons; they are vexatious to the spirit. (Max Ehrmann)
- gaf1958
- Trusted Seller
- Posts: 12431
- Joined: Sat Jun 16, 2012 6:24 am
- CW-watches: 24
- Location: ɐᴉlɐɹʇsn∀ 'ʇsɐoƆ ǝuᴉɥsunS
Precision, Rate and Accuracy in watches: a layman's view
That was both an enjoyable and educational post Steve. Thanks for taking the time to put it to words, with a touch of your usual flair of course. It took some things I already knew, but tied them together in a way that made me understand the concepts better - so thank you indeed.
CW C1+2xC3+6xC6/60K+C7+C11+3xC60T+2xC65+C90+2xC600
Omega Ω 11xSpeedy+14xSeamaster+4xConnie+DeVille
Cartier+2xPanerai+2xFarer+2xOris+Sinn+11xSeiko+ManyVintage
B&R+Halios+5xVisitor+TagH+6xTissot+2xZelos+4xCertina+more
Family12xCW+2xΩ+Cartier
Omega Ω 11xSpeedy+14xSeamaster+4xConnie+DeVille
Cartier+2xPanerai+2xFarer+2xOris+Sinn+11xSeiko+ManyVintage
B&R+Halios+5xVisitor+TagH+6xTissot+2xZelos+4xCertina+more
Family12xCW+2xΩ+Cartier
-
- Similar Topics
- Replies
- Views
- Last post
-
- 10 Replies
- 297 Views
-
Last post by Noush
-
- 6 Replies
- 548 Views
-
Last post by Bahnstormer_vRS
-
- 52 Replies
- 1848 Views
-
Last post by TheBeatles
-
- 8 Replies
- 406 Views
-
Last post by ajax87
-
- 4 Replies
- 580 Views
-
Last post by cward210