closing tag is in template navbar
timefactors watches



TZ-UK Fundraiser
Results 1 to 20 of 20

Thread: A Small Timekeeping vs Cost Experiment

  1. #1
    Grand Master TaketheCannoli's Avatar
    Join Date
    Feb 2011
    Location
    UK
    Posts
    18,851

    A Small Timekeeping vs Cost Experiment

    Does the cost of a watch dictate the quality of the movement and accuracy of its timekeeping? Should it?

    I think it's a reasonable assumption that there should be a correlation between price and accuracy. I was in a conversation about this recently and it was really obvious that the non-WIS strongly believe this to be the case. I explained that sometimes a £150 automatic watch can comfortably out-perform a £3000+ watch, a comment that was met with derision and not a small amount of mocking.

    So I thought I'd do a small experiment within the £2k price range - I'm not made of money!! So, did I fall on my backside by being completely wrong and retreat into my cave?

    I bought / tested four watches costing between £334 and £1850, all brand new. I set and fully wound them (all got 50 winds) and tested their timekeeping over three durations of 24 hours having worn them indoors in the same way and all were left dial up[ overnight.

    Purchase #1

    Tag Heuer Aquaracer with Calibre 5 movement at a cost of £1850.

    Performance: gained between 17 and 25 seconds per day.


    Purchase #2

    Squale 1521 with ETA 2824-2 at a cost of £799.

    Performance: gained between 12 and 17 seconds per day.


    Purchase #3

    Timefactors Speedbird III with Sellita SW200-1 'Top' at a cost of £485.

    Performance: gained a consistent 5 seconds per day.


    Purchase #4

    Seiko SRPE99 PADI Turtle with 4R36 a cost of £344.

    Performance: no deviation.

    I obviously didn't expect that set of results! I must say even after all my years of owning automatic watches that this is a surprise. I'd suggest that only the Speedbird performed as expected. The TAG was a disgrace. I got the salesman patter about "you're buying a Ferrari of the watch world here". Honestly the whole buying experience was nauseating. I was embarrassed for them.

    The Seiko on the other hand could be mistaken for a quartz.

    You can make of this what you will but I hope you at least found it interesting.
    Last edited by TaketheCannoli; 26th July 2021 at 14:26.

  2. #2
    Master
    Join Date
    Dec 2010
    Location
    NW Leics
    Posts
    8,153
    It is interesting. Quartz watches can also defy expectations; for example I have Casios costing less than £20 that are substantially more accurate than both my solar Seikos with RRP in excess of £200.

  3. #3
    Grand Master
    Join Date
    Mar 2006
    Location
    Somerset (U.k )
    Posts
    12,254
    Blog Entries
    1
    What an interesting experiment.

    Did you fully wind each watch before the 24hr timing period?

    I wonder what the use of a Timegrapher would show straight out of the box.

    as for the results, well I’m not entirely surprised to be honest. We all know that accuracy is a very hit and miss affair but on paper you would expect a higher cost watch to perform better than a cheaper one.
    Cheers,

    Ben



    ..... for I have become the Jedi of flippers


    " an extravagance is anything you buy that is of no earthly use to your wife "

  4. #4
    Master
    Join Date
    Dec 2010
    Location
    NW Leics
    Posts
    8,153
    Come to think of it my most valuable quartz watch (a midsize 007 SMP) is one of my least accurate. But I've had it since 1998 and I think that's possibly a consequence of its age.

  5. #5
    Master
    Join Date
    Apr 2015
    Location
    Stockton, Teesside, UK
    Posts
    1,496
    Quote Originally Posted by monogroover View Post
    It is interesting. Quartz watches can also defy expectations; for example I have Casios costing less than £20 that are substantially more accurate than both my solar Seikos with RRP in excess of £200.
    Well yes. The REAL Timekeeping vs Cost experiment is to pit a £15 Casio quartz against a range of ludicrously expensive mechanical watches, and wonder why you wasted your money on them!

  6. #6
    Grand Master Sinnlover's Avatar
    Join Date
    Sep 2014
    Location
    London
    Posts
    10,945
    I think for me the eye opening fact is that the TH, TF and Squale all have the same basic movement in them, yet time keeping differs so much. There is also a huge price differential between them.
    Spending the big bucks does not mean you are guaranteed a better watch unfortunately, there are lots of other things that bump the cost up. Some tangible, some less so.
    Last edited by Sinnlover; 26th July 2021 at 10:58.

  7. #7
    Grand Master TaketheCannoli's Avatar
    Join Date
    Feb 2011
    Location
    UK
    Posts
    18,851
    Yes all were fully wound.

    Quote Originally Posted by ben4watches View Post
    What an interesting experiment.

    Did you fully wind each watch before the 24hr timing period?

    I wonder what the use of a Timegrapher would show straight out of the box.

    as for the results, well I’m not entirely surprised to be honest. We all know that accuracy is a very hit and miss affair but on paper you would expect a higher cost watch to perform better than a cheaper one.

  8. #8
    Master Papa Hotel's Avatar
    Join Date
    Oct 2012
    Location
    Not Edinburgh
    Posts
    7,487
    Meanwhile, my Turtle loses around 25 spd and my Panerai, with a RRP of just over 10 grand is 0 over a week.

  9. #9
    Grand Master Saint-Just's Avatar
    Join Date
    Apr 2007
    Location
    Ashford, Kent
    Posts
    28,935
    There was a time when accuracy was a selling point. If you read some of the achievements by brands like Zenith at the beginning of the 20th century it is astounding.
    Unfortunately it is not so much the case now, and regulating a watch takes a little time that many brands do not want to spend, regardless of how much they charge.
    All the mechanical watches above can achieve much better (bar the Seiko, although we would need to know if the watch was worn at all time, or spent time in one position and time in another, etc. to form an opinion; still great result, though).
    'Against stupidity, the gods themselves struggle in vain' - Schiller.

  10. #10
    Grand Master
    Join Date
    Jul 2007
    Location
    Wakefield, West Yorkshire
    Posts
    22,498
    Quote Originally Posted by TaketheCannoli View Post
    Yes all were fully wound.
    Are you sure?

    This is a very odd set of results and the first thing I’d question is whether they were all in the same state of wind initially. An ETA 2824 takes approx 35 twists of the crown to fully wind, Seikos take around 60, Miyotas take 60, an ETA 2894 takes around 60.

    Suggest you repeat this, but wind them all 60 twists of the crown. Also, take readings at intervals, this will show how the rate is changing as the watch unwinds.

    Think about what this really tells you, it tells you how the watch runs over 24 hrs in the dial up position. Think about what’s happening to the rate between 18 and 24 hrs, a watch with a 38 hr power reserve will be running at below 50% power reserve, that’s going to affect amplitude and likely to affect rate.

    The only meaningful test is to either wear each watch for several days or simulate this by running on a cyclotester for 16 hrs then allow 8 hrs overnight dial- up.

    Comparing the watches on a timegrapher in all 6 positions is a far better reflection of the differences between them. Compare at full wind, compare after 12 hrs running, compare after 24 hrs, this will give a far clearer picture.

    Whichever way you look at it an ETA 2824 or Sellita is a better movement than a Seiko, but the differences can easily be masked by differences in regulation.

    Repeat this exercise as I’ve suggested and see what it tells you.

  11. #11
    Grand Master Neil.C's Avatar
    Join Date
    Sep 2003
    Location
    SE England
    Posts
    27,035
    Quote Originally Posted by TaketheCannoli View Post


    Purchase #4

    Seiko SRPE99 PADI Turtle with 4R36 a cost of £344.

    Performance: no deviation.
    A bit simplistic to test one of each model.

    Also they were probably not all fully wound and in all positions as they would be when worn.

    FWIW I've had a couple of the 4R36 Turtles and they were +20 and +30 respectively out of the box.
    Cheers,
    Neil.

    My Speedmaster website:

    http://www.freewebs.com/neil271052

  12. #12
    Grand Master TaketheCannoli's Avatar
    Join Date
    Feb 2011
    Location
    UK
    Posts
    18,851
    In response to a couple of statements. This was a bit of fun and never intended to be a scientific experiment. I've updated the original post to describe how I tried to align the environments as much as possible by giving them the same number of winds, wearing them the same way and leaving them in the same overnight position.

    I can't believe that doing it very differently would make huge differences.

  13. #13
    Master
    Join Date
    Aug 2012
    Location
    N. Ireland
    Posts
    1,158
    Quote Originally Posted by walkerwek1958 View Post
    Whichever way you look at it an ETA 2824 or Sellita is a better movement than a Seiko, but the differences can easily be masked by differences in regulation.
    Very much this. What the OP is observing is how poorly regulated some movements can be and is in no way a reflection on quality of Sellita vs Seiko vs ETA movements.

    In my experience a new SW200 or 2824 can be regulated to within +2/-2 seconds over six positions, but I don't think any manufacturers bother with that as it's time consuming and costs them money, so when you buy a new watch and discover it's running 10s fast the AD can say 'it's within specification sir.'

    As for the Seiko 4R36, spec for those is +40/-30, but you can sometimes get lucky and find one thats near as damn perfect on the wrist. Over the last couple of months I've regulated over 100 brand new NH35's (a 4R36 for the commercial market) and the first 100 had an average variance in rate, or delta over six positions of 12.9s. The best ones had a delta of 4s and there were only three of those out of the 100, three had a delta of over 40s and deemed unusable and two were DOA.

    So yeah, it's not impossible that a humble Seiko can outperform a watch costing 10x the price, but it's very much luck of the draw.

  14. #14
    Grand Master Onelasttime's Avatar
    Join Date
    Oct 2014
    Location
    Everywhere, yet nowhere...
    Posts
    13,717
    I can't see any meaningful info gained from this test?

    If any new ETA 2824 was giving +17-25 per day I'd be sending it back for regulation or refund, despite the basic 2824 having these parameters, or near enough. I'm sure the Calibre 5 has better figures than this on paper so send it back. Same with the Squale.

  15. #15
    I enjoyed reading about your experiment TTC and it’s the sort of thing I would have done to amuse myself some time ago. My Ocean X and Seiko PADI Turtle are easily the most consistent timekeepers in my collection. I’ve just checked my Ocean X and it is +20 seconds against time.is and was set over a month ago. That’ll do for me.

  16. #16
    Grand Master
    Join Date
    Jul 2007
    Location
    Wakefield, West Yorkshire
    Posts
    22,498
    Quote Originally Posted by Dynam0humm View Post
    As for the Seiko 4R36, spec for those is +40/-30, but you can sometimes get lucky and find one thats near as damn perfect on the wrist. Over the last couple of months I've regulated over 100 brand new NH35's (a 4R36 for the commercial market) and the first 100 had an average variance in rate, or delta over six positions of 12.9s. The best ones had a delta of 4s and there were only three of those out of the 100, three had a delta of over 40s and deemed unusable and two were DOA.

    So yeah, it's not impossible that a humble Seiko can outperform a watch costing 10x the price, but it's very much luck of the draw.
    V. interesting! I`ve handled several fairly new ones where the owner reports problems and I've ended up stripping and rebuilding the movement, invariably I`ve been able to get them all to perform well but sometimes it's required some serious intervention. I`ve seen some which have been over-oiled and some where the lubrication appears to have been missed in key areas. Others have required the hairspring to be manipulated and trued up to get good performance. Very much a case of luck of the draw, these movements are capable of excellent performance but they are quite sensitive to state of wind too, so it's not unreasonable for different owners to get different results owing to variation in wear pattern.

    Seikos aren`t the easiest to regulate, it's nigh-on impossible to change the rate by 1 or 2 secs/day because there's no fine regulation, you end up losing the setting and starting again which is v. frustrating. If I`d regulated over 100 in a couple of months I think I would've ripped my hair out. Miyotas are the same but not quite as bad, it's a shortcoming in the design and maybe that's part of the reason they aren`t regulated well from new. I`ve also come across significant beat errors on newish Seikos, that just requires more care in production/adjustment.

  17. #17
    Grand Master snowman's Avatar
    Join Date
    Nov 2012
    Location
    Hampshire
    Posts
    14,536

    I have to agree, a TAG or a Squale running over +10 seconds a day brand would seem to be performing badly, but I'm not sure what level the ETA2824-2 is as the standard is rated +/-12 seconds per day up to +/- 30 sec/day, so performing in spec if it's that in either watch.

    In an equally non-scientific test, my most accurate watch on arrival since I bought a timegrapher was a £60 Kommandirski!

    It recorded +/- 0 s/d!

    I've bought a LOT of NH3* movement watches over the last couple of years - Worst was +20s/d, best was +2s/d, most have been sub 10s/d.

    M
    Breitling Cosmonaute 809 - What's not to like?

  18. #18
    Grand Master MartynJC (UK)'s Avatar
    Join Date
    Dec 2008
    Location
    Somewhere else
    Posts
    12,336
    Blog Entries
    22
    In my experience there is NO correlation between price and accuracy.

    I have had 20K+ watches running at +15sec per day. I’ve had £2K watches run with COSC. Work horse movements like the 3135 run for years just fine but don’t have any fancy finesse finishes (hence no case backs). ETA 2824 etc produce v accurate movements.

    BUT - it’s not a rule. Both my VC and Patek keep good time
    Last edited by MartynJC (UK); 27th July 2021 at 13:41.
    “ Ford... you're turning into a penguin. Stop it.” HHGTTG

  19. #19
    Master
    Join Date
    Aug 2012
    Location
    N. Ireland
    Posts
    1,158
    Quote Originally Posted by walkerwek1958 View Post
    V. interesting! I`ve handled several fairly new ones where the owner reports problems and I've ended up stripping and rebuilding the movement, invariably I`ve been able to get them all to perform well but sometimes it's required some serious intervention. I`ve seen some which have been over-oiled and some where the lubrication appears to have been missed in key areas. Others have required the hairspring to be manipulated and trued up to get good performance. Very much a case of luck of the draw, these movements are capable of excellent performance but they are quite sensitive to state of wind too, so it's not unreasonable for different owners to get different results owing to variation in wear pattern.

    Seikos aren`t the easiest to regulate, it's nigh-on impossible to change the rate by 1 or 2 secs/day because there's no fine regulation, you end up losing the setting and starting again which is v. frustrating. If I`d regulated over 100 in a couple of months I think I would've ripped my hair out. Miyotas are the same but not quite as bad, it's a shortcoming in the design and maybe that's part of the reason they aren`t regulated well from new. I`ve also come across significant beat errors on newish Seikos, that just requires more care in production/adjustment.
    I've been working for a local micro brand doing warranty repairs which is basically new movement swaps because they're so cheap and not worth servicing, but I've checked a few of the fauly movements and found a couple where the impulse jewel on the balance had dropped out, some had no end shake on the balance so very low amplitude and poor timekeeping, and hairsprings not bent correctly is another common one.

    If you check the technical guide for the 4R/NH it says to oil the pallet fork pivot jewels which no other manufacturer states you should do. Even older Seiko tech guides say not to oil them so I'm not sure why they're doing it now. They also state you should oil the flat side of all pivot jewels instead of the usual oil it from the oil cup side and I've no idea what they're thinking there.

    Spending days at a time only regulating them has been an excercise in boredom and frustration; Make an adjustment, sit for 30s and wait for the rate to stabilise, make another adjustment and wait 30s and repeat endlessly. I've found that they can't be adjusted in less than around 10s increments due to the friction between the stud and regulation arms, so if I've got one that needs to be +3s/d dial down and it's reading +7 I need to move the regulator arm too far in one direction and then move it back again and hope it lands on +3, which of course it rarely does. I'd say over 90% need the beat error adjusting but I never recorded any numbers for that. The majority are within 1.0ms but there is the odd one running at over 2.0ms and 9 times out of 10 the stud needs moved in the same clockwise direction.

    The owner of the company got fed up with complaints of poor timekeeping etc so now every watch is regulated by me before it's sold. I'm currently doing around three per hour and I want a couple of extra timegraphers so I can do three at a time and not have to sit waiting for 30s after every adjustment, but ultimately I'd like to see a move to Sellita movements for ease of regulation and also improved performance and reliability. In the meantime I have over 400 NH35's waiting to be regulated...

  20. #20
    Grand Master Neil.C's Avatar
    Join Date
    Sep 2003
    Location
    SE England
    Posts
    27,035
    Quote Originally Posted by MartynJC (UK) View Post
    In my experience there is NO correlation between price and accuracy.

    I have had 20K+ watches running at +15sec per day. I’ve had £2K watches run with COSC. Work horse movements like the 3135 run for years just fine but don’t have any fancy finesse finishes (hence no case backs). ETA 2824 etc produce v accurate movements.

    BUT - it’s not a rule. Both my VC and Patek keep good time
    Assuming that high quality movements that are adjusted to five positions including isochronism will cost more than cheaper unadjusted movements I think there is a very definite correlation between price and accuracy.

    Any movement can be regulated to a certain extent, although of course it is far harder to regulate a cheaper balance (as mentioned earlier) compared to the better quality micro type but a fully adjusted movement will keep far superior time overall than an unadjusted type.

    It's why they are more expensive.
    Cheers,
    Neil.

    My Speedmaster website:

    http://www.freewebs.com/neil271052

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

Do Not Sell My Personal Information