Sooth Ski: Comparing Skis with Measured Data (2100+ skis!!)

[alude]

Having these ski measurement comparisons available will be helpful for picking ski models and lengths to try, similar to how looking at geometry charts and suspension kinematics curves for bikes can be used. But I have found with bikes that what’s on paper can give a starting point, and there are certain elements that can turn out differently when actually riding (some of that could come down to the accuracy of the available information).

Yeah, definitely. It should be taken as a starting point and we are curious to see how far we can take it.

I think one of the question that comes the most often is if a specific ski length actually skis short/long (compare the next length, or to a different model). I think we could do a decent job at answering that by presenting measured length, running length, mass, and stiffnesses. That seems to be useful for hard snow. Anything else that would be useful? Right now, I am trying to figure out how to present that information efficiently.

I think we could also do something like that for float. I think in that case we should consider surface area, mount point, tip height and bending stiffness. Anything else?

I will be interested to test how on-snow feel compares with what I would expect from the ski measurements.

I think this is the ultimate challenge. Looking at someone's review and finding a way to explain that in the data is somewhat easy. The real challenge is to figure out what are the most important things to consider when comparing two skis.

[alude]

I've been examining the Sooth data to scrutinize those claims by alude.
CONCLUSION: Great job at measuring stiffness, Sooth!

Thanks for spending the time to check the data. We are still working on your list of improvements. Should be coming soon.

Vitamin I - if you could somehow get all the skis from your collection into the DB, I would be immensely grateful!

Might happen someday. Most of my collection is in Truckee, so Sooth could bring their machine for one trip to measure:
- my skis,
- Moment factory
- Praxis factory,
- Mountain Mike's shop at Olympic Valley carries a bunch of maggot-worthy lengths of Palisades-worthy models (at least last time I checked, before COVID era)

First, we would need to help alude figure out:
- what time of year is best for the most inventory at Moment factory, Praxis factory, etc.?
- how to confidently subtract the "binding effect" from measurements of my skis, and from the demo-mounted skis at Mountain Mikes, etc. (because I ain't willing to unmount all my skis for this project).
.

I am starting to understand the interest Marshal! Real cool project! Happy to contribute if we can.

It would be really nice to do a measurement run in California/WestCoast for personal collections, other brands and shops. A few years ago we measured about 500 skis from the Vermont Ski and Snowboard Museum (mostly wood skis) and 200 from the personal collection of a friend in Ontario (part of the results are here: https://www.mdpi.com/2504-3900/49/1/135, we were trying to document the history of skiing). Vitamin-I has quite a collection!!

As far as bindings, it is best if we remove them but I think there are ways to work around that.

[Vitamin I]

RE: Interpolation and rare weird tails (e.g. swallowtails, etc.)

...
Similarly, [for Sakanas] this link below shows "incomplete horizontal sweep" of the sensor for the region at the end of the tail, even for a 2022 ski. This "sweep without any load" omitted the tail width point, and so results in bad consequences for many other reported values.
https://compareskis.shinyapps.io/com...=%22Compare%22

...
As much as I like skiing the Sakana, they are a b**** to measure. The measuring head fall off the swallow tail before we can complete the width measurement (similar problem as with convex bases). I think one year we patched the hole with tape so that we could get better measurements (2021).

I am realizing that the interpolation of height/camber doesn't work here though. Usually we do it only for the last 5 cm so it creates very little error. But in this case, interpolating over 20 cm just doesn't work. Any suggestions on what to do instead of interpolating to still show where the end of the ski is?

Given that the Comparator graphs are already displaying both measured data as well as "interpolated data", one idea is to display solid line for ONLY the measured data, and display a different style of line (like dashed line) for other displayed data (like for interpolated data). If users can become accustomed to that convention, then it would be acceptable to use ANY non-measured method for the dashed line regions, including a subjective "artist's depiction". I think an artist's depiction for the end of the Sakana tail would be more accurate than an attempt by automated interpolation/extrapolation. And an "artist" could just search the web for photos of Sakana, and the artist could eyeball the height/camber profile at the end of the Sakana tail in the photos, or count pixels or whatever.

.

[cmor]

Some of you have asked for it, so here is a little bit more about the history of Sooth Ski. (e.g., having a machine in each shop to measure the wear of skis... but turn out that stiffness doesn't change with use so that was out, offering ski fitting sessions, etc.).

I'm surprised there's no difference. I've come back to skis after a break, that have a few hundred days on them and they were not what I remembered. How heavily used were these beatered skis?

[alude]

Given that the Comparator graphs are already displaying both measured data as well as "interpolated data", one idea is to display solid line for ONLY the measured data, and display a different style of line (like dashed line) for other displayed data (like for interpolated data). If users can become accustomed to that convention, then it would be acceptable to use ANY non-measured method for the dashed line regions, including a subjective "artist's depiction". I think an artist's depiction for the end of the Sakana tail would be more accurate than an attempt by automated interpolation/extrapolation. And an "artist" could just search the web for photos of Sakana, and the artist could eyeball the height/camber profile at the end of the Sakana tail in the photos, or count pixels or whatever.

.

Yeah, I think it is the best idea. It will take a little while to implement (double the amount of lines you need to manage), but we will do it.

[alude]

I'm surprised there's no difference. I've come back to skis after a break, that have a few hundred days on them and they were not what I remembered. How heavily used were these beatered skis?

I am surprised too. I think we all experienced ski ageing!

We did a few tests:
- Flex a ski for something like 100,000 cycles on an instron machine... but that was only in bending, maybe not enough amplitude and maybe not enough cycles.
- Measured the SL and GS skis of a U18 racer throughout the season... that was last year, so the racer didn't get as many turns as in a usual season due to covid. Maybe we should have done it for longer. He told us he could generally feel a difference between his skis (to justify buying new skis every year). Well, the way he described it was that he thought he would be able to feel a difference after many days of skiing on each of them... not during a quick A/B blind tests. I think some of it comes from learning to trust a pair more than the other during races and skiing more confidently, not really feeling a difference during a turn.

In both cases, we didn't measure any difference beyond the 5% accuracy of the machine. Unfortunately, not measuring a difference doesn't really mean that nothing was different or that nothing would eventually become different. We just can't tell.

However, we talked to a few composites expert and they told us that the stiffness typically doesn't change much in a layup as the fibres of the composite are still all there and doing their job. However, I think it might be slightly more complex than that (you can look for residual stiffness of composite in google)...

Other things can also change, including damping (e.g., from cracks in the resin) or camber changes (e.g., permanent shift between the layers).

I am also not talking about very cheap skis (e.g., skis from the 90' with foam in them, very soft skis, etc.).

[alude]

We added ski overall taper angle to the list of specs...

We also investigated with Vitamin-I the change in camber in our measurements. It is a little bit long to fully describe, but the key takeaway is that camber measurements will be affected by 1 to 4 mm if gravity is playing a role (i.e., either upside down or right side up). The exact value depends on weight, stiffness and distance between supporting points. Best thing would be to measure skis while they are sideways...

[alude]

FYI, we changed the way we deal with flat/reverse cambered skis. A few flat cambered skis were being identified as reverse cambered due to the local curvature around the mountpoint. This triggered a different calculation of camber for reverse cambered skis. This was not a very consistent or logical way of doing things (i.e., what is the camber of a reverse camber ski?).

Anyways, we do not calculate the camber of reverse cambered skis anymore. They will be listed as 0 mm of camber (minimum possible value). This strategy should also produce much more consistent results for flat cambered skis, and better reflect what we can see on the profile view of the ski.

[margotron]

just fiddled around. will say this is dope

[alude]

I was talking about skis on another forum, and we found something interesting about the Monster 108 (and other Monsters). I thought you guys might like it.

Yes it is a heavy ski, and yes it is stiff in bending for it length. But what is really interesting is how stiff the tip/tail are in torsion (2-3x) while the center part is very soft (75% ish). See below or on our website to compare to other skis.



This torsional stiffness distribution is pretty unique to Head skis and found on some of their racing skis. I am not sure what it is supposed to do, but it would be surprising if it wasn't intentional. Maybe that explain why so many people find something special about this ski?

[TheK12]

Alude - How are you measuring torsional stiffness? I played around with your ski comparison tool (which is really cool by the way) and there are some numbers that I found suprising. For example, the 2021 Armada Tracer 98 (an all mountain ski) 180cm has avg torsional stiffness of 108 while an Elan Wingman 82 CTI 178cm has an avg torsional stiffness of 106? Am I to read this as the Tracer having better edge grip than the Elan Wingman, which is by all accounts a carving ski? Am I mis-interpreting the data or missing a normalization that accounts for ski width? And in that same comparison, the Armada Declivity 92ti is off the charts at avg torsional stiffness of 141? I would have expected the carving ski to have a much higher avg torsional stiffness than the wider all mountain skis, even the ones that contain metal.

And finally... why "sooth ski". Trying to figure out origin of that name.

[alude]

Alude - How are you measuring torsional stiffness?

In short: we twist the ski and measure the base twist angle at each mm along the full length of the ski. Then we differentiate to get the change of twist angle for a given length, and use that and the applied torque to calculate the torsional stiffness (aka GJ). There are obviously a few more tricks to make that work, but that is the general idea.

I played around with your ski comparison tool (which is really cool by the way) and there are some numbers that I found suprising. For example, the 2021 Armada Tracer 98 (an all mountain ski) 180cm has avg torsional stiffness of 108 while an Elan Wingman 82 CTI 178cm has an avg torsional stiffness of 106? Am I to read this as the Tracer having better edge grip than the Elan Wingman, which is by all accounts a carving ski?

First, I assume you mean the 2022 Elan Wingman 86 CTI 178cm. We don't have the 82 in that length in the database.

These two skis are challenging to compare (here are the details in the comparator). It is much easier when the skis are closer to each other. In this case, the tip/tail of the Tracer and Wingman are shaped differently, the weights are different by almost 600g and the widths are also different by 12 mm. When you compare skis that are closer to each other on most other aspects (and that happens more often than you might think), you are right to think that torsional stiffness is a very good predictor of the edge grip that you will feel on snow.

To really answer your question, we need to think about what is edge grip. Honestly, it is not fully understood, and it might mean a few different things to different people. We are working on a better definition(s). But ultimately, the way I think about edge grip is having the edge stay in its tracks. To improve that, you can:
1. Distribute your weight uniformly on a longer section of edge so that you won't break the groove in the snow and start sliding
2. Prevent the ski from twisting and sliding out of the groove
3. Prevent the ski from flapping around at high speed, cause it will ruin #1 and #2
4. Resist twisting loads at the tip to better dig a groove in hard snow
5. Etc.

Many factors are to be considered for each of these to happen. Any of them could destroy the edge grip of a ski. Shape is super important for #1, while bending stiffness, camber and mass are super important for #3 (if you ski fast). Torsional stiffness is mostly important in #2, but also #3 and #4.

The Wingman is heavy and is stiff in bending. That means it speed limit will be higher than the Tracer (#3). The running length of both ski are close, but the Wingman has a very shallow rocker with less tip taper (see below). This means that as soon as you tilt the ski, more and more of the edge will engage the snow (#2). It is the same thing with the tail.




Now, if you look at the average numbers for torsional stiffness, the Wingman seems to be about 20% softer (note: torsional stiffness differences between skis can be as high as 200-400%, so 20% is not that high). However, if you look at the detail graph you will find that the torsional stiffness distributions of both skis are almost identical, except underfoot. This is probably a binding reinforcement. We still have to confirm that, but have indications that the torsional stiffness under the foot is not super important (this is related to how a beam flex/twist). Anyways, if you discard that region, the two skis are identical in torsion.




So both skis have similar torsional stiffness, but the Wingman is stacking up a few features that would make me think that it would hold an edge better on snow than the Tracer.

That being said, its torsional stiffness level is medium for a ski of that shape. You can read Blister's review (https://blisterreview.com/gear-revie...wingman-86-cti), specially the section "who is it for". They use a lot of words to says that it is an intermediate level ski. They say it doesn't quite match the edge hold of most piste skis, that you should use it if you find other skis unforgiving, that is it not for ex-racers, etc.

Am I mis-interpreting the data or missing a normalization that accounts for ski width?

We are working on tools to consider all these factors together. Otherwise it is just too confusing. We recently discovered that normalizing torsional stiffness by width and effective edge length make the comparison of torsional stiffness easier between different geometries. We still have to fully test this, but it seems to make sense physically. Torsional stiffness is there to resist the torque on the ski. The wider the ski, the more torque you will create. The longer the ski, the more it will twist.

That would further explain why I would not say so quickly that the Tracer would have a higher edge grip (i.e., it is wider for the same torsional stiffness).

And in that same comparison, the Armada Declivity 92ti is off the charts at avg torsional stiffness of 141? I would have expected the carving ski to have a much higher avg torsional stiffness than the wider all mountain skis, even the ones that contain metal.

The Declivity and the Wingman are much closer in shape (Declivity has more tip/tail rocker) and weight (but still, 300g difference) than the comparison with the Tracer. They also have similar bending stiffness, and the torsional stiffness are starting to be more different in this case (40% on average).

If you have access to Blister's deep dive, you can compare these two skis (careful: Blister compares the 184 Wingman to the 180 Declivity). They talk about the Declivity being looser and easier to pivot (due to its rocker shape) but the Wingman being more forgiving (lower torsional stiffness). We can only take all that with a grain of salt though, because we haven't measured the 184 Wingman.

One thing that you might like to compare is how different is the Wingman CTI vs the TI (here). The CTI is rated as a mid-advanced to high-expert ski (https://elanskis.com/eu/wingman-86-cti-fusion-x), while the TI is high-intermediate to low-expert (https://elanskis.com/ca_en/wingman-86-ti-abdhby21). The CTI is supposed to have carbon tubes in the core to increase torsional stiffness. Well, see below the measurement for two lengths of 86 CTI/TI.




Torsional stiffness is not really higher for the CTI version. The CTI with "Tubelight Woodcore" are also not really lighter. I would be really curious to see if these skis would really feel different on snow in a blind test! Maybe there is something that we are not measuring?!?

I am pointing that here for the Wingman, but really you can find that stuff from most/all compagnies. Many like to talk about torsional stiffness, but measured data is important.

And finally... why "sooth ski". Trying to figure out origin of that name.

Sooth means "truth" in old english (unfortunately we found that it is not something well known, and only french speaking people could discover such a thing after spending too much time looking in an english dictionary around a few too many beers)! It also comes from sooth sayer, which is a "person supposed to be able to foresee the future". It is also related to soothe, which can mean "to bring comfort, solace, or reassurance to", "to bring peace, composure, or quietude". Hopefully our measured data will be another tool in the bucket to help skiers know what they like and buy skis with confidence that they will like them.

[ARL67]

^^^ Thanks for the detailed explanations using the above examples, very informative as always !

[TheK12]

^^^alude - Thank you for that awesome explanation! And yes, it makes sense that edge hold is not just a function of torsional stiffness but is also a function of ski shape and bending stiffness. Ultimately, I love that you guys are providing this data! It's very "soothing".... (Groan... very much pun intended... ). Thanks again!!!!

[abraham]

Torsional stiffness is not really higher for the CTI version. The CTI with "Tubelight Woodcore" are also not really lighter. I would be really curious to see if these skis would really feel different on snow in a blind test! Maybe there is something that we are not measuring?!?

Maybe something else related to spring behavior like the natural frequency? I'm not sure how easily you'd measure it with the setup you have if I'm remembering the explanation correctly.

[alude]

Maybe something else related to spring behavior like the natural frequency? I'm not sure how easily you'd measure it with the setup you have if I'm remembering the explanation correctly.

Natural frequency is a function of both stiffness distribution and mass distribution. We just measure the total mass. It is possible that these two skis have different mass distributions.

We don't measure natural frequencies, mode shapes, modal damping, etc. We can do it, but it takes too much time to apply to 1000+ skis per year and it is hard to relate these measurements to the actual behaviour on snow.

[powbmps]

Did Sooth Ski disappear? Website no longer seems to exist.

[alude]

It should only be temporarily.

You can still compare skis there in the meantime: https://compareskis.shinyapps.io/compare

[powbmps]

It should only be temporarily.

You can still compare skis there in the meantime: https://compareskis.shinyapps.io/compare

Awesome! Thank you.

[alude]

Back online.

We just started measuring 22-23 skis. Should be all in there within 3-4 weeks...

[DerKommissar]

If I can ask, where do you get the skis from?

[alude]

Many shops graciously allow us to measure the skis in their inventory. They enjoy having our measurements as it helps them in helping their customers. We do a few different trips near Quebec where we are based.

Any manufacturer can also send us their skis. We will measure them, free of charge, add them to the comparator and return them unharmed. This is good for direct-to-customers brands, brands that are not well represented in shops in the North East, etc.

[alude]

We have added about 300 2022-23 skis to the comparator: https://soothski.com/compare/

We measured 350 more at Ski Essentials last weekend. We just need to process them. They should be added to the comparator within 1 week.

We should do 2-3 more measuring sessions. Let us know of any special requests and we will see what we can do...

[J. Barron DeJong]

Let us know of any special requests and we will see what we can do...

Heritage Lab

[alude]

Heritage Lab

Ahaha, of course! Happy to measure any skis that Marshal send our way and return them unharmed.

But with most of Heritage models being already sold out, I am not sure if that will happen.