Maybe have one of them be a more traditional frame/disc-brake combo? Canondale or Parlee? I see All3 in Atlanta sent up the P5 - they (well, Podium) seem to have been building up a number of the Parlees lately. Maybe they could volunteer one for round 2.

^^^This^^^
But I don't think anyone cares about such an antiquated bike with all the high dollar super bikes out now.

I'd like to see an aero road bike with aero bars such as a S5. See if one bike can do it all, or how close it actually comes.

---

--Chris

This tested an aero Venge ViAS vs. a Tarmac and found a 5 minute difference over 40k. That's a huge number but I'm really not surprised. I think after a bike is aero optimized is when we start to see diminishing returns.
https://www.outsideonline.com/...found-will-shock-you

Edit: 2 minutes over 40k with frame to frame comparison. The other 3 minutes seems based on wheels, shoes, helmet, skinsuit. They also say the Venge ViAS is as aero as the SHIV TT. So I think can a lot can be extrapolated from that.

---

-------------------
Madison photographer Timothy Hughes | Instagram

There it is again.

Why is it antiquated? Just because the frame design is 5-6 years old? Or is it because it is notably slower than frames released in 2017?

One of those claims, if true, has merit. The other one does not.

The Shiv has upright dentist geometry. I would have to make miracles happen to fit on that bike.

Because there have already been tests that compare the Shiv Tri to the P5 and others and it was nothing to write home about.

Why?

Somebody really, really, really needs to test a Fuji Norcom Straight (with fast handlebars).

I'd be up for that.
I'd also like to test the Orbea Ordu
diamondback Serios which I've heard good rumors about but haven't seen hard data

the problem with Fuji and Orbea is their dealer networks suck and that makes them non relevant to the triathlon world.

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

Where are we with Scott's Plasma?

IIRC, you had some pretty good results on that bike back in the day.

Now, the Brownlees join SK to fly the flag.

Seems we need to know the truth about the modern Plasma.

Scott

I was wondering the same thing. It looks to be a good set up with everything one would need at a great price point.

For me, I'd like to see:

Giant Trinity
Canyon Speedmax - Ultegra Di2 version
Scott Plasma - new Tri version a la Brownlee
SC
QR6

Felt IA (interesting take on the double diamond)
New BMC (it's new and probably performs similarly to the Canyon et al.)
TR Omni (novel design from small company)
SC 7/Giant ("Budget bike")
P4 (how much progress has been made since 2010)

Argon 18 E-119
Canyon Speedmax
Giant Trinity
Scott plasma 5
Cannondale Slice

---

res, non verba

Rude. My dentist rides a P4!

---

/kj

http://kjmcawesome.tumblr.com/

What I'd like (but it's probably a few years away at very best) is a CFD program that takes information from a camera or LIDAR suite and then does the work for you.

---

'It never gets easier, you just get crazier.'

iirc the new plasma's are 7-9w faster than the old plasma 3. With some equipment mods from stock you can cut that gap in half...if not more.

I think Scott suffers from the same problem as Fuji & Orbea. Great bikes crappy consumer education program and crappy dealer network. I wouldn't even know who to reach out to get my hands on a new Plasma here in the US. I probably have a better network for this sort of thing than the avg triathlete

---

Brian Stover USAT LII
Accelerate3 Coaching
Insta

If some, like Dan's, predictions are correct, and all the frames in the first test are within say 10 watts of each other, which would possibly be within a margin of error too? Is it really worth taking a bunch more frames in? Seems like you'll have simply proven that there just isn't much difference between all the top frames, and folks should concentrate on other things (position....).

I'd much rather see the same sort of unbiased test done on wheels. Guessing you'd see larger differences.

Been there, done that...it does ;-)

---

http://bikeblather.blogspot.com/

I wouldn't think that +- 10 watts would be the margin of error. If this were the case WT testing would be pointless. My guess is that we are reaching the point of minimum drag for a bike, though something like the Vias Venge demonstrates that some crazy design work and singular focus (I.e., good low-yaw performance) can still save 5-10 watts.

Jim at ERO made the point that the next step in aero is to determine the best rider/bike combo. Nice idea but I don't see this as practical. It is hard enough to find a few aero helmets, in the correct size, to test. Imagine the $ needed to provide a range of bikes.

After my DA was stolen I did this with an AR using a Tririg alpha x and reversed seat post. Damn thst was good looking and light - probably 16.5-17 pound range. Ultimately I got tired of the swapping back and forth to go for Saturday morning group ride and ended up building an IA 10 with the alphaX front end. I picked that frame based on this advice from Desert Dude in this thread:

http://forum.slowtwitch.com/...IA_14%3F_P6044198-2/

This is a question I've been thinking about a lot. What is the difference between a "source of error" I need to just explain and one I should quantify. In other words, what is industry best practice for error bars here?

I can calculate SEM a number of different ways.

Watts?

With a pedaling cyclist on board, CdA is generally reproducible to within +/- about 0.004 m^2, or +/- 2%. Differences smaller than that could still be real, but you'd need multiple measurements to be sure. (OTOH, differences larger than may still not be real, but odds of that go down the bigger the difference found.)

Was the drag of the fixture measured separately?

If so, I would recommend the approach that I used when testing brakes.

If not, it's a bit of a moot point.

I don't know if they ran it with only the fixtures. I asked that same question and didn't get a clear answer. Will try to find out. I really hope they did, but it's not obvious to me from the data I have that they did.

In terms of the "moot point", I don't know about that. I had thought of using the control vs. the baseline to establish variability from those runs. This assesses "variability from human rider moving" error, which people were way concerned about in one of the other threads.



Maybe this doesn't make any sense -- I don't know. Anyways, now that I've tidied up the data and removed all the hard coded figures in the calcs, I'm ready to hand it off to an adult for review.