Haven't seen exact measurements of the 5000 on a FC rim (though I'm sure they're out there), but I would assume the safest option would be 23mm up front and 25mm rear like you suggest. The 5000 is such a great tire when you factor in everything that you really don't need to overthink it there. Following Josh's "rule of 105" I would imagine the 23mm inflated would come up a hair narrower than the rim which is what you want.

---

Benjamin Deal - Professional - Instagram - TriRig - Lodi Cyclery
Deals on Wheels - Results, schedule, videos, sponsors

That data of ours was from a Zipp 858 - on an 808FC I would run either a 23mm GP5000 or 23mm Corsa Speed if you want to risk it. Or in between would be a 23mm GPTT which is a touch faster than the 5000.

---

AeroCoach UK
http://www.aero-coach.co.uk

https://blog.silca.cc/...ure-and-aerodynamics was an illuminating and highly relevant bit of work, thanks to both of you for pushing my research in that direction. I was vaguely aware of the 105% rule before but had not expected the resulting aerodynamic effects near that boundary to be as significant as the chart showed. 25mm would appear to have just over 1W Crr advantage per wheel for me at target speed, while the aero penalty compared to 23mm at anticipated pressures and widths would probably be more like 2-3W per wheel at relevant yaw angles, if I'm interpreting the graph correctly. Granted, I'm sure the effects are muddier at the rear wheel, but it looks like I'll be ordering a 23mm set.

Thanks again!

Zipp recommends 23 front 25 rear for firecrests.

Hi Carbenfire, If you really want to overthink it, then you can project the exact CRR vs aero differences for the Conti5000 23 vs 25 vs 28 (or indeed any tire on the market) correcting for your weight and typical speed here: http://fft.tips/tyre2 (enter "normal" into the free text box to force recommended inflation pressure). Of course there are several assumptions in a calculator like this, it will never tell you your exact personalized losses but it will get 95% there. bw Alex

Alex, fascinating resource, thank you for sharing. Very powerful for quickly interpreting results from such a comprehensive compilation of data. I'm excited to see how closely it matches with my efforts from gathering the same information last night... can of worms is slowly opening back up again.

At 23mm/Latex/167lbs combined/45kph the raw cost for a GP5000 shows as 58.27W, while increasing to 25mm yields a result of 61.17W. A 3W difference for the wheelset is almost exactly the net result from what I described above and seems to further validate my choice.

One question, however. I also compared these two against the 25mm TL version with the same parameters and saw that the resistance dropped further to 55.57W. This was somewhat surprising, as I was under the impression that tubetype clincher + latex was roughly equivalent to the matching tubeless setup--or perhaps I'm thinking of results for running latex in the TL variant? Taking rolling resistance in isolation, I see that this accounts for most of the difference (as expected), with 36.7W for the tubetype and 30.4W for tubeless. This is somewhat surprising even after normalizing the values against Xavier's test results referenced above to account for varying protocols and show proportional differences. There, the tubed tire w/latex had a (scaled) 1.6W advantage vs. a 6.3W disadvantage as displayed on the spreadsheet.

I'm not sure which database the latex datapoints came from after a quick scan through each of the ones linked, but this variance seems roughly in line with the butyl numbers provided on BRR.com. Scaling from his 18 mph single wheel tests to a pair at 45kph is roughly the same as tripling the values, so taking the closest run for each (80psi) yields 12.1W*3 = 36.3W for the butyl tubetype clincher and 9.9W*3 = 29.7W for the TL version, compared to 38.9W (latex) and 32.7W, respectively, for isolated rolling resistance values from the spreadsheet at weights that recommended 80psi in each case. I don't know if normalizing w.r.t. the TL result is appropriate in this analysis, but the end result seems to have the spreadsheet imply that latex only has a roughly 1W advantage over butyl at 45kph, which seems wrong. In Xavier's other test the difference between the same tubes is measured at closer to a 7W advantage, which is almost exactly the difference described above and consistent with generally accepted results.

Is it possible that this is mislabled and actually represents a butyl-tubed test instead of latex? Again, not sure, since I couldn't find the source, but the numbers appear to indicate the difference pretty clearly. Or perhaps it's just a coincidence that the testing differences are in the same ballpark as tube differences.