Why not make a "soft catch" mode for Grigris for potential factor 2 falls?

Eric Carlos wrote:

This video also doesn't represent the technique in a soft catch, because the belayer is static, they are just adding slack to the system.  That is not how a soft catch works.

Edit: In later tests, they do a better representation, but only slightly.

It's pretty clear here that you haven't actually watched the video.

Edit: In response to your edit - could you explain to me what is wrong with their soft catch technique, or methodology?

Alexander Blum wrote:

It's pretty clear here that you haven't actually watched the video.

That is why I went back to finish it.  Their early examples were terrible.  But I agree that a soft catch is a good thing in most circumstances.  

Eric Carlos wrote:

That is why I went back to finish it.  Their early examples were terrible.  But I agree that a soft catch is a good thing in most circumstances.  

When would a soft catch be bad then?

Gumby King wrote:

When would a soft catch be bad then?

That catch should match the terrain, the gear, the belayer/climber weight ratio, and a number of other factors that most gumbies cannot understand.  But hopefully with a good in person tutor, they can move beyond gumbyland and figure these things out.  

Eric Carlos wrote:

That catch should match the terrain, the gear, the belayer/climber weight ratio, and a number of other factors that most gumbies cannot understand.  But hopefully with a good in person tutor, they can move beyond gumbyland and figure these things out.  

Please speak Gumby.  So when is a soft catch bad?

Depends on Terrain?  You dont say

Climber/belayer weight ratio?  Does this matter for the "depends on terrain"?

Other factors!?  

Can you mentor me?  I have much to learn. Because no additional insight was shared this must be complicated.

Gumby King wrote:

Please speak Gumby.  So when is a soft catch bad?

Depends on Terrain?  You dont say

Climber/belayer weight ratio?  Does this matter for the "depends on terrain"?

Other factors!?  

Can you mentor me?  I have much to learn. Because no additional insight was shared this must be complicated.

You have already stated that for you "Soft catches are overrated in the outdoors unless you are climbing some bad ass overhang route."  Stick with your gym climbing and trolling.  

Artem V wrote:

A soft catch will reduce the speed of the climber before they come in contact with the wall - a hard catch won't do this and more deceleration will occur when the climber comes into contact with the wall (this rapid deceleration is what causes the injury). You will always have flight time before you touch the wall on a vertical wall (less than overhanging) - the only time you wouldn't is if you are on a slab (less than vertical).

I would love to demonstrate this using a haul bag dummy and an instrument to measure the force of impact.

As an analogy - think of a dynamic line vs a static rope. Which would you rather fall on?

Another analogy: What sounds more appealing - falling onto a device attached to a fixed anchor or onto a belayer 20 lbs lighter than you? Which will be more painful? A hard catch will trend you more towards the fixed anchor/static rope - a well timed hop will trend you more towards the light belayer/dynamic line.

Artem, I figured out the disconnect. You're taking "less than overhanging" to ONLY mean plumb vertical. The rest of us are taking "less than overhanging" to mean, well, all climbs that are LESS THAN OVERHANGING. That means everything from some 5.2 beginner trad route to big portions of even El Cap.

Alexander Blum wrote: No, I didn't miss that point. I looked up the video because I essentially agree with you from a physics perspective, but it doesn't align with my personal experience. The portion of the video where the climber is 0.5 m from the pendulum point models a vertical, not overhanging, route pretty well (IMO).  My ass is often 1.5 feet from the wall when climbing, and if I fall doing something dynamic I would imagine it could end up out even farther than that. The video concludes - through mathematical analysis and empirical modeling - that climber speed as they impact the wall is lower with a soft catch in that situation. The data shown when demonstrating this convinced me that the same would be true on a slightly slabby route. 

Here's what's missing in the video, IMO: She's falling into empty space. Put an 85 degree angle of granite under her and redo it. Then let's talk. 

Alexander Blum wrote: Artem,

This youtube video demonstrates your point pretty well.

Haha, especially at the end when they say if you are doing trad climbing, forget everything we said. (But, actually,  some of it should be relevant...)

Their results are very interesting and one has to appreciate the thought and work that went into the video. It introduced a new concept to me (new to me the first time I saw the video), which is the potential for the rope contracting elastically and whipping the climber into the wall on the bounce back from maximum extension, something that can indeed be mitigated by a properly-timed soft catch.  Proper timing is critical, and their description doesn't match what is often recommended, since their model suggests the properly-timed jump begins just before the rope is tensioned---obviously a  tricky thing to gauge and impossible to get right when the leader is out of sight.  This point is driven home at several moments in the video, when he says maybe he didn't time his jump right, and by the model-produced graphs below, which illustrate the substantial variation in outcomes that result from different jump timings. These different timings are all within a bit less than half a second, in other words the jumper has to nail a sweet spot inside a half a second target to have a beneficial effect.



Something to understand about jump timing is that a belayer has no way to improve with practice unless there is technical data providing feedback about the impact results, something that no climbers other than the two in the video actually get.  Chances are a lot of climbers who think they're giving a soft catch and think they are experienced at it aren't managing to consistently time it within that 0.5 second sweet spot to produce a positive as opposed to negative or no effect, because they've never had the kind of technical feedback that would enable them to condition their behavior.  

Something similar happened years ago with the original dynamic belay, which involved letting a certain amount of rope slip through the belay under tension.  Eventually tests were formulated and carried out that showed that experienced people who had practiced a lot weren't able to consistently reduce peak anchor loads.  I think this might have been one of the driving factors in the production of ropes whose behavior was sufficient, or close to sufficient, to arrest falls without special belayer interventions.  My guess is that the jump soft catch may be prove to be analogously inconsistent in its results, and I might add that the level of belayer attention one sees nowadays at gyms and sport crags makes the kind of split-second precision timing needed for an effective jump catch highly unlikely in general.

Practically speaking, I'd guess almost all climbers activate later than the ideal moment, when the rope tension begins to be felt. This won't result in the maximum possible reduction in impact, but will probably have some positive effect.  Lighter climbers don't have to do anything, as they'll be pulled up without reacting.  Heavier climbers will have to master the timed jump.
 
All this is in addition to the unclear effects of a jump soft catch on less than vertical walls that we've been discussing.  In the case when the climber started a half-meter out from the "wall," it seemed to me that the climber's feet crossed the imaginary wall line a little before the belayer jumped (check me on this at 8:55--8:59 in the video), meaning the belayer's reaction was not going to reduce the spiking effect much or at all.  Perhaps with the wall relatively close, by the time the rope tension has built up to cue the belayer to jump, the climber's feet have already struck the wall.

Another consideration is that the jump might only be capable of reducing the energy the rope has to absorb by a relatively fixed amount; it may not scale with the fall height (as happens with an elastic medium like the rope), and so might become less and less effective the further the leader falls, because the fixed amount of energy it can absorb becomes a progressively smaller fraction of the total fall energy.  The tests in the video never have the faller above the high gear point, so we don't know how effective the jump is as the fall length increases.  Moreover, in a real situation with the faller above the top piece, they arrive at the starting point of the video with an initial velocity, not the zero velocity of the tests.  How this initial velocity alters the subsequent trajectory we do not know, but we can be certain the results will not be the same.  I presume the mathematical model they have could give insight into this.

A lot of this is moot for multipitch trad anyway, since the reality of trad stances and belay situations means that the jump catch won't be possible.  It is looking more and more as if a Munter hitch directly on the belay anchor may be the most effective belay---assuming the anchor is good enough for this.  And some relatively recent tests done by ENSA in a gym with sport-climbing type falls found that a Munter hitch on an anchor on a wall performed better at reducing impacts than the usual jump belay.

An additional concern has to do with the methodology of the video, which I think calculated velocities from screen data.  In order for these to be accurate, the swinging climber has to be moving in a plane precisely parallel to the picture plane.  I don't recall hearing any a of possible errors due to possible lack of parallelism and no error intervals around the measured values were provided.  They didn't seem to have the time to do multiple trials that could be averaged either.  This means there is some uncertainty in the interpretation of differing measurements.

Dave K wrote:

Even plumb vertical isn't going to change the parameters much. Once again, rgold nailed it: "on less than vertical ground, [or even plumb vertical] the pendulum never crosses the vertical"

Of course on plumb vertical there could be some pendulum as we do tend to jump out a bit when we fall, and our center of gravity could be a couple of feet from the wall when the fall is initiated. But it's not enough of a pendulum that it needs to be offset at the expense of a longer fall and more likelihood of the climber's foot glancing off a feature.

This debate is basically spray vs. science. On one side there are people that think they are expert belayers because they understand nuance and complexity that does not actually exist. On the other side there are those that understand the relatively simple physics involved in determining the trajectory of a lead fall.

No need to make it personal. I'll be the first to admit I'm not versed in the physics, but in my experience, on a vertical wall I much prefer a moderately soft catch if there is nothing to hit (I suppose we could debate ad nauseam what constitutes a "feature," but I'm thinking of climbing in the 5.10/11 and up range, so fairly small holds. A ledge, even a small one, would change the calculus. We could also debate ad nauseam what constitutes a "moderately" soft catch). Most of my partners agree, and that's the kind of catch I will give them, or anyone in such a scenario unless they explicitly request otherwise. The alternative--a hard lock-off--results in, not necessarily a "spike," but at the least, uncomfortable impacts on vertical walls.

On a long trad pitch with the leader up high there's so much rope out I'm not worried about this at all either as the belayer or climber. I'm thinking more of single-pitch, one-rope, vertical sport and trad. Since I suck at overhangs but still like to push my limits when I can I probably have more experience on this terrain than anything else, so that's where I'm coming from here. Call this spray if you must but no one posting in this thread--not even RGold!--has given me any real-world reason to reconsider this approach. 

Gumby King wrote:

When would a soft catch be bad then?

When the extra length of the fall would cause the leader to hit a ledge or the ground or other feature.

"There is an old quarrel between philosophy and poetry" -Play-Doh

So a munter then...
Well at least it’s free.

Grigris kill.

Figuring out a “send” mode for climbing shoes sounds just as possible and far more useful.

rgold wrote:

Haha, especially at the end when they say if you are doing trad climbing, forget everything we said. (But, actually,  some of it should be relevant...)

Their results are very interesting and one has to appreciate the thought and work that went into the video. It introduced a new concept to me (new to me the first time I saw the video), which is the potential for the rope contracting elastically and whipping the climber into the wall on the bounce back from maximum extension, something that can indeed be mitigated by a properly-timed soft catch.  Proper timing is critical, and their description doesn't match what is often recommended, since their model suggests the properly-timed jump begins just before the rope is tensioned---obviously a  tricky thing to gauge and impossible to get right when the leader is out of sight.  This point is driven home at several moments in the video, when he says maybe he didn't time his jump right, and by the model-produced graphs below, which illustrate the substantial variation in outcomes that result from different jump timings. These different timings are all within a bit less than half a second, in other words the jumper has to nail a sweet spot inside a half a second target to have a beneficial effect.

Something to understand about jump timing is that a belayer has no way to improve with practice unless there is technical data providing feedback about the impact results, something that no climbers other than the two in the video actually get.  Chances are a lot of climbers who think they're giving a soft catch and think they are experienced at it aren't managing to consistently time it within that 0.5 second sweet spot to produce a positive as opposed to negative or no effect, because they've never had the kind of technical feedback that would enable them to condition their behavior.  

Something similar happened years ago with the original dynamic belay, which involved letting a certain amount of rope slip through the belay under tension.  Eventually tests were formulated and carried out that showed that experienced people who had practiced a lot weren't able to consistently reduce peak anchor loads.  I think this might have been one of the driving factors in the production of ropes whose behavior was sufficient, or close to sufficient, to arrest falls without special belayer interventions.  My guess is that the jump soft catch may be prove to be analogously inconsistent in its results, and I might add that the level of belayer attention one sees nowadays at gyms and sport crags makes the kind of split-second precision timing needed for an effective jump catch highly unlikely in general.

Practically speaking, I'd guess almost all climbers activate later than the ideal moment, when the rope tension begins to be felt. This won't result in the maximum possible reduction in impact, but will probably have some positive effect.  Lighter climbers don't have to do anything, as they'll be pulled up without reacting.  Heavier climbers will have to master the timed jump.
 
All this is in addition to the unclear effects of a jump soft catch on less than vertical walls that we've been discussing.  In the case when the climber started a half-meter out from the "wall," it seemed to me that the climber's feet crossed the imaginary wall line a little before the belayer jumped (check me on this at 8:55--8:59 in the video), meaning the belayer's reaction was not going to reduce the spiking effect much or at all.  Perhaps with the wall relatively close, by the time the rope tension has built up to cue the belayer to jump, the climber's feet have already struck the wall.

Another consideration is that the jump might only be capable of reducing the energy the rope has to absorb by a relatively fixed amount; it may not scale with the fall height (as happens with an elastic medium like the rope), and so might become less and less effective the further the leader falls, because the fixed amount of energy it can absorb becomes a progressively smaller fraction of the total fall energy.  The tests in the video never have the faller above the high gear point, so we don't know how effective the jump is as the fall length increases.  Moreover, in a real situation with the faller above the top piece, they arrive at the starting point of the video with an initial velocity, not the zero velocity of the tests.  How this initial velocity alters the subsequent trajectory we do not know, but we can be certain the results will not be the same.  I presume the mathematical model they have could give insight into this.

A lot of this is moot for multipitch trad anyway, since the reality of trad stances and belay situations means that the jump catch won't be possible.  It is looking more and more as if a Munter hitch directly on the belay anchor may be the most effective belay---assuming the anchor is good enough for this.  And some relatively recent tests done by ENSA in a gym with sport-climbing type falls found that a Munter hitch on an anchor on a wall performed better at reducing impacts than the usual jump belay.

An additional concern has to do with the methodology of the video, which I think calculated velocities from screen data.  In order for these to be accurate, the swinging climber has to be moving in a plane precisely parallel to the picture plane.  I don't recall hearing any a of possible errors due to possible lack of parallelism and no error intervals around the measured values were provided.  They didn't seem to have the time to do multiple trials that could be averaged either.  This means there is some uncertainty in the interpretation of differing measurements.

If this post from a Mathematics University Professeur (also with his long-time practical experience in climbing which is many more years than most of us have been alive) does not end this discussion on "soft-catch" (and he even acknowledges that under ideal circumstances it might just be better) then there is no way that you are open to changing your mind.

Thanks RGold for taking the time!

Didn’t someone ask years ago, “Why can’t belay devices have an idiot mode?”  And thus the Gri Gri was born.  

Don’t push your luck. Lol. 

As someone who's climbed a lot of steep stuff in the Red but also a lot of vert and slab in NC, I almost always give and prefer to receive a soft catch. I don't know what people are talking about when they say you should give a hard catch on slab. I've taken plenty of slab falls, and the ones that suck the most are when I get a hard catch. On everything but very low angle friction slab you're usually falling slightly away from the wall, and a hard catch increases the impact into the wall. On vertical terrain I couldn't imagine thinking a hard catch is better, you're falling away from the wall and you're going to swing back in so it'd be dumb to make that swing fast and short with a hard catch. Honestly, I sometimes prefer a hard catch on really steep stuff because I'm not gonna hit anything but air anyway, and it makes it less work to boink back up. 

The Grigri 7

There are many ways a climber can get spiked on a 'vertical' route. Admittedly, probably not as hard as on an overhung route, but it can still happen.  Climber could fall while doing a move that requires lots of outward force such as a layback or pulling on an undercling with smeared feet.  Another scenario is anytime you're not directly above your last pro such as on a traverse.  Traversing out of a corner would essentially be the same as a roof.