Back to the topic at hand… I am not convinced (yet) that a 2.1 or larger (MTB) tire is faster for every possible gravel event than the best 45-50 mm (Gravel) tires. But I am going to do my own testing. The Schwalbe G One RS has been my choice of tire for racing since it has been available, first in 40 mm, then in 45 mm. This tire is screaming fast, and is one of the absolute fastest gravel tires available by any and all accounts. My own testing last season proved to me that the 45 is the faster choice in every situation… My next test will be the Schwalbe Thnuderburt, 2.1, super race, if I ever get them.
I do not think anyone here is advocating for bigger is always better, after all, is anyone really considering racing gravel on 2.6 DH tire… probably not. But there is a sweet spot, and that sweet spot is looking like it is with a larger tire than has traditionally been used for gravel, after all, it was not that long ago when some thought a 40 was getting big! Now I am certain there is no reason to ever run a 40 for any gravel course, beyond that, I am not sure, it will seem to depend.
I also do not see the need to be testing tires that are not fast already, in any given size. The Pathfinder is really stout and not supple, and it is slow, perhaps new versions will be better… Pirellis, not so good, Maxxis, not so good, why bother even considering these slow tires (regardless of size). I was a bit disappointed by Dylan’s Silverstone testing, considering he used Pirellis as his benchmark “gravel” tire, and added to that he used tubes in the Pirellis and sealant in the Race King…
The Pathfinder Pro 42mm is in the top third of tires tested by BRR 40mm and larger. 42mm or larger it’s exactly the median tire. It’s not fast but it’s not that slow either. I think it makes up for it in general durability. It’s also probably the most common or in the top 3 most common gravel tires since it’s release. Knowing where it stacks up would be worthwhile to a lot of people.
For instance, the WTB Resolute was one of the only mid-range 700cx42mm tire for a couple years and I thought they were great, felt fast and rode well. Then they hit BRR and turns out they were pretty slow and getting slower every year in comparison. Knowing this helped me evaluate newer tires and judge if it was worthwhile to look closer and upgrade.
Fast isn’t the only thing. If it was, the René Herse Extralights would be a lot more popular.
The 700cx48 Thundero tests pretty well, I got a flat my first event on it that knocked me off the podium. Never had a flat with Pathfinders…
There don’t seem to be many 45-50mm tires worth really looking at that aren’t already tested at BRR.
René Herse 700cx48 - the RH 700cx44 is pretty good in both regular and EL.
Gravelking X1/R/SS 700cx45 - the 40mm model tested slower or about like the Pathfinder so maybe not?
Schwalbe G-One Speed SG 700cx50 - the 40mm didn’t test very well either.
What else is there?
Hiya. What test method is this? I’m doing the G One RS tire you mentioned soon and excited to see what shakes out
That fundamental issue is what keeps me using pathfinders. Is a 5w faster tire a faster tire if you have to dart it?
Just ride my normal training loops at a constant power. I ride the same loops a lot of times. So, just look at the segment times after, and average up the results from the long term. I also throw out any days with significant wind. These rides are 2+ hours, and about 75% gravel or so. Then I look at Strava segments to compare things like times on pavement sections, etc. This testing is real world, and not accurate enough to consider small differences as significant, but it does show enough difference to make some relatively accurate inferences.
I stand by my statement. I feel that I can choose the widest and fastest tire for the race to give my self the best chance of landing on the podium. BRR is a great tool to help you select tires, but it should not be the only tool used to determine what tire you should use.
thundero running a 48
terra speed running a 45
g-one 45
not running pirelli
pathfinder 47/42 they both measured with in 1mm when on rim
getaway pro 45
strade bianca 40
I can not see a smaller tire that is better then the bigger one.
Thanks for the extra info.
Why not test the new G One RS Pro 50mm tire as well as the Thunderburt 2.1? I suspect the G One would provide better puncture protection and might be worth sacrificing a couple of watts.
I’ve just had my first ride on a set of Hutchison Caracal Race tires and wow are they fast. Mixed surface, probably 70/30 gravel/road. I had a 4x20 sweet spot workout outside today and set several pr’s as well as a couple KOM’s without even trying that hard. I hit speeds I’ve never seen on my gravel bike.
It was wet out today and the grip was as good as any other tire I’ve tried. They do let go sooner than something with knobs however.
On my wheels, they measure out to 43.5mm at 31psi. I think these are my new go to race tires.
Might snag some Burts and Chung them against each other to test this threads premise. I’d be shocked if a 2.1 Burt could be faster than these. I’ve been shocked before though. 
Agreed, I am very interested in that comparison. I will probably pick up a pair of the G One RS in 50 mm at some point as well. I have the Thunderburts on B/O, but who knows when/if they come available.
Doing those soon. Should be a good showdown!
Looking forward to your results.
I listened to your podcast today (the free version), which was enjoyable. There were a couple of things that came to mind while listening, which might be useful for yourself or anybody else doing their own Chung testing for gravel/MTB tyres:
You mentioned that the Chung Aerolab analysis in Golden Cheetah has limits on the CdA and CRR sliders. That’s true, and those limits have caused me some frustrations in the past. The guy who coded it, Mark Liversedge, kindly modified GC a few years ago after I suggested extending the range of CRR values. Even with the newer GC versions though (I’m using v3.5), the maximum CRR of 0.015 isn’t high enough for soft surfaces, as you commented on the podcast. However, even though the slider stops at 0.015, you can type larger CRR values into the left-hand box and the virtual elevation chart will take that higher value into account - it’s just that can’t use the slider.
Secondly, there have been a few comments in this thread about the effect of wider tyres on CdA, with the concern being that the reduced aerodynamic efficiency of wider tyres might offset some or all of their CRR benefits. CdA assumptions were a point discussed in the podcast. The usual way that Chung testing is done is by fixing CdA and the using it to determine CRR, or vice versa, depending on whether somebody is doing an aero test or a rolling resistance test. However, if you know that both CRR and CdA will change, as is the case when comparing a narrow tyre and a wide tyre, it can be quite tricky to determine both. It’s still possible to figure out which tyre is faster overall though, and relatively easy. What I do is fix the CdA value for both tyres (but accept that the fixed CdA assumption is incorrect). Then, when doing the Chung analysis, any reduction in CdA will be attributed to an reduction in CRR. The split between CRR and CdA won’t be correct, but that often doesn’t matter, because we normally just want to know if tyre A is faster than tyre B, considering both rolling resistance and aero combined. It’s important to do the test at representative race speeds though. You can even fix both CRR and CdA in the Chung analysis, and then the setup whose virtual elevation profile descends the most is the fastest tyre.
Thanks for the detailed reply!
As far as manually inputting the CRR range… WOW. It sure does work. Probably for the best, as I ended up talking with Dr Chung over that issue and what he taught me on how to calculate I have actually had to do here and there. One of those times, a mistake came out good in the end! I am using 3.6 Aerolab, which goes to 0.025 now with the slider which is pretty sweet. That’s cool you talked to Mark, I would like to also at some point.
I think the testing almost goes in levels of difficulty. Seeing changes in CdA is easiest. Finding absolute CdA values is harder than that as you need to know Crr then. Then would be finding changes in CRR with known CdA as you start really needing to know elevation profile. I like what you suggest and have done it that way here and there…it fits nicely in with the rest of it. The hardest as you mention is pulling CRR and CdA apart from each other in one test.
Robert Chung talks about this some in the main pdf he wrote on the testing. What he suggests is variable speed uphill runs, and I modified that a bit. It works, but you have to do everything correctly and have to have just the right course profile…meaning shallow gradient so you can vary speed and power enough climbing to actually get a delta. The analysis is also tricky, I have been referencing topo maps to know true elevation change, and basically am able to find the CRR and CdA separation between the high speed and low speed climbing laps. I can usually get them to better than a half meter change per lap and essentially there is only one CdA and CRR combo that will get the change correct between laps.
It is total overkill for most uses of Chung method but for this, it just seemed like the most accurate way.
The CdA penalty for let’s say…a 45 vs 61 mm tire do exist, but in terms of wattage are a third of amount you get back in improved rolling resistance off road as you mentioned.
For anyone who thought the thread was getting boring, we probably either made it or ruined it totally with this exchange, haha!
Cool. I’ve had some success varying speed to try to tease apart CdA and CRR (also based on Roberts’s advice!), but it sounds like you’ve developed much more advanced techniques I’ve used.
Yeah, apologies to anybody following this thread that just wants to be told what tyre to run.
On the contrary, I think that’s the whole ball game. There can be no definitive answer to the OP without having a grasp what both CdA and CRR are doing. Weight too for that matter. It’s a complicated equation. I love the details, keep them coming!
For anyone who thought the thread was getting boring, we probably either made it or ruined it totally with this exchange, haha!
Way better, of course. Because this is what this thread should be all about (or at least, way more about). Improving ones understanding of tools and methodologies and results instead of “yeah but in the real world…”. Or the kind of bitching it gravitated towards especially the last days (totally unnecessary and reminded me of very early internet days).
To the topic at hand: I haven’t run into slider boundaries with the Chung Aerolab Tool in GC yet. But that’s just because of my test course surfaces. Good to know that you could enter values even beyond the boundaries set by the slider. That said - I mostly never touch the sliders anyways but type in my values in the box anyways.
And indeed, because of the way the resistance terms are (with different powers for v in the cda and crr parts) you can derive both crr and cda via one run if you vary the speed in it enough. But it’s also true: it’s a finicky analysis to do. There’s a reason one of the papers I cited waaaaaaay early in this thread analyzed every test run they did by a range of different persons.
I also always have to do it “by hand”. The Aerolab has this “Auto” labeled (or something along those lines) button but you obviously can’t use it for out and back style test runs (because of braking, turning etc.). Has anybody in here a classical loop available as test stretch? Like Chung describes in his og paper? One where you don’t ever have to brake and can run loop after loop. Wonder if you could use the Auto button there to any success…
I have a loop. Still don’t use auto button in GC for it.
Holy cow, I didn’t know you could just type in a value above the slider range.
I only ever test road tires so I’m learning a lot by hearing these tips and tricks about testing MTB and gravel tires. Thanks, both to you and to @jkarrasch !