Adarq.org

I have a challenge (probably impossible) for Adarqui (I would also like vag to chime in or anybody else for that matter):

Here's the deal: Someone at work asked me if something is mathematically possible:

One company records its invoices totals with numbers with 4 decimals.

Say you buy 3 items that each costs 1.0250 $. The total would be 3.0750 $ (with 4 decimal numbers).
However, this company calculates the item prices with only 2 decimals.

So they would calculate 1.02+1.02+1.02 and expect to find 3.0750 at the end. Our company has two choices (according to this retarded company) - either truncate (and end up in the 1.02x3 situation) or round (and end up in the 1.03x3 situation).

Both, as you can see, are "wrong". One gives a total of 3.06 (instead of 3.0750) and the other gives a total of 3.09 (instead of 3.0750).

This, for many products, leads to a significant different between the calculation and the actual total.

I told them it's impossible mathematically, since if we "cut" the last 2 decimals we're basically giving up that information for good, and creating an approximation. And I guess that's that and end of story.

I tried coming up with something... weird... probably idiotic, although I tried it a bit and it was more precise in some situations.

Here's the algorithm that I thought about:

I basically established a "weight" for each item - say you have 10 items. 4 of them are of type "last 2 decimals under 50", 6 are "last 2 decimals over 50".

Like this:

10.1241
10.4126
10.8744
10.5616

Again, the last 4 decimals all "under 50".

Then the other 6 (chose 6 randomly) would be this:

10.5589
10.2377
10.3497
10.1166
10.4351
10.7888

Then I calculated the sum in 4 decimals (the total, that they want in 4 decimals).

SUM4 = 104.4595

Now the sum if we're to truncate the last two decimals

SUMTRUNC = 104.40

Now the sum if we're to round the last two decimals (if under 50, keep the 2 decimal of the number unchanged, else increase by one):

SUMROUND = 104.46

Now I thought, what if I pretend all the last digits are UNDER 50 for one calculation, and all the last digits are over 50 for another, and then assign weights to them?

So that would be SUMWEIGHT = 104.40*0.4 (because only 4 elements actually have last 2 decimals under 50) + 104.50 * 0.6 (we're assuming that we round up all the numbers, but only 6 elements have last 2 decimals over 50) = 41.76 + 62.70 = 104.46 -> which ends up exactly as the rounded sum.

I did several calculations and it was different in different scenarios. There's also the possibility that my calculation was wrong, or that I'm doing something completely redundant or idiotic to begin with.

I mean, the more I look at it the more it seems like I'm doing a more complicated rounding, basically (which is probably exactly what I'm doing) - do you have any other ideas or this is simply an impossible task?

interesting point about the hip displacement. i *feel* like i'm trying to take my hips back, so not sure how i could load the hips more without dropping them, i.e. bending knees. but maybe i should do that. something to try.

Well, hip hinge, just like in a RDL or a KB swing. Should be easier to do, since you have the weight of the med ball in front of you for additional balance.

There's tons of resources... www.codecademy.com , https://www.hackerrank.com/domains (linked by Andrew above), and also bitspyder.org as a tracker for tutorials (tons of them) if you manage to get an invite for an account (I did so years ago).

Otherwise... google, lol

What you're describing sounds somewhat similar to what's happening here, too. We didn't even had a teacher of computer science in high school - nobody would come to teach at the school for a salary of 200$ per month in this domain.

Plus, I absolutely hated programming with all my heart, all my life. Started badly, was "teached" poorly, and I have always thought to myself "these programmers brag about how smart they are. If they're so smart, how come they need 1000 lines of code to calculate two numbers??? Why does it have to be that complicated, with so many words written, to calculate two numbers?".

And based on that, I absolutely hated programming. This continued in college too - I would not go to the classes and play ball instead. And we did Java and PHP in college, I barely attended the classes and was never interested in what was in them, couldn't understand a thing. It was basically the teacher writing weird stuff on the blackboard, continuously, and that was it, you go home. Nothing to understand.

Obviously it was very complicated for me back then too, as my mom and dad would break dishes and swear at each other and threaten each other and had to go in ambulances with my mom and emergency hospitals and crazy, crazy, crazy stuff, while also being in love like crazy on top of it all.

In the class, by the way, people would smoke and play poker, and throw papers at the teacher (in high school).

So... yeah... it is what it is.

There's so much to write about, lol.

No, my bet would be on the viscosity of the joints and stuff like that. Being very plyometric and very weak, that's of paramount importance. Probably after all that warmup the body temperature reaches a point where it's optimal for "plyometric work" so to speak. from the standpoint of structural factors (viscosity, tendon properties and stuff like that).

He should have advocated 100% blood flow restriction for no less than 5 minutes while performing as many 10RM reps as you can within those 5 minutes. Actually no, even as a joke that's deadly. HEH!

Let me send that as a challenge to Donald Drumpf on twitter. brb

Yeah back to the hip thrust argument - there's such a big variation into what people "consider" to be a hip thrust. I was watching LBSS's videos of his hip thrust, and he uses such short ROM. When I do the hip thrusts, I go to full hip hyperextension (I try, at least), pause there for a second or two, then go all the way down. I don't go 1 cm and then back up, I go all the way down so the plates touch the ground, and then do the exercise again. Plus sometimes I feel the exercise in the lowerback, meaning I wasn't actually using my glutes properly/I wasn't getting into posterior pelvic tilt, I was trying to "back extend" - so the completely wrong movement, and also with injury potential.

Sometimes I like to just touch the ground with the plates and go straight back up, so that I maintain continuous tension in the glutes for the entire set, sometimes I rest the barbell on the floor for a second and do the next rep (maybe I'm repositioning my back on the bench etc). But there's a tremendous amount of difference in TUT (or the total amount of time it takes me to do say 5 reps) vs what I've seen around here, with 1-inch humps of the bar being called "hip thrusts". Certainly a lot less time spent per set for the same amount of reps, a lot less stretching of the glutes, a lot less range of motion etc.

This makes me think about chinups or squats or whatever - when you ask someone how many chinups they do, they might say "15". When you tell them to show you how they do them, they do 2 inch chinups (similar to 1/4 - 1/8 squats that you see the regular gym idiot do).

So yeah, it seems that we have completely different understanding of what a properly performed hip thrust is. And yeah, there's definitely variation going on depending on how you position yourself.

I remember doing the hip thrusts with Olympic lifting shoes on (I think LBSS did so, too) and they felt totally different than the same exercise without OL shoes on. I even reported that to Bret Contreras some time ago (I think he said he hasn't studied the difference).

Also, you can kind of push through the "toes" (or the front part of the foot) or through the heels. Personally, I feel weird pushing through the heels. You can also kind of pull the ground back with your feet while in that position (although that would activate the hamstrings as knee flexors, especially when the bar is resting on the floor and the hamstrings are not in active insufficiency).

So I agree, there's a ton of stuff going on with the hip thrust. I think properly performed KB swings would do a lot of good for maxent, as would depth jumps for height with a target straight overhead (this will force him to hip extend a bit more, IMO, than a regular DJ that for him would be so quad-dominant).

Quote from: Raptor on April 05, 2016, 04:03:20 pm

When he gets some glutes? Wasn't he saying that he's hip thrusting 200+ for reps? Doesn't sound like lack of glute development to me.

The problem with using an exercise to gauge power.  He can't jump more than 30'' and he is incredibly slow.   Watch his dunk videos and you see a guy who has undeveloped glutes that don't fire.  Hip thrusts are a good glute activator but they use a small range of motion and the load doesn't necessarily equate to power very well.  I can get a good workout hip thrusting 225lbs or 505lbs - I have never "maxed" my hip thrust and don't need to.  One of the biggest evaluation mistakes you can make (and usually it's self evaluation where we make it) is when we assume that an athlete can do "X" so therefore that part of the body is clearly not the problem.   It usually comes from an athlete doing an isolation exercise of limited range of motion and assuming that means that muscle is not an issue.  Classic example is a slow sprinter with weak hamstrings who can do something silly like a "Natural GHR" assuming that "I don't have weak hamstrings so that can't be the issue" but not realizing that using your hamstrings as a knee flexor with your knees on the floor doesn't have much to do with hamstring power or development in hip extension during a sprint.  Watch an athlete move dynamically an locate where he has weakness rather than referring to a gym lift as evidence of anything.   

I agree with that. But you can't say he "has no glutes". You can say he isn't using the glutes that he already has at a pretty good level in terms of strength (yes, isolated movement strength, gym work strength, whatever) well in dynamic movements. That I can agree with. On the other hand, he doesn't seem too fast in ANY movement so I'm not sure it's the glutes' fault - it's either his desire to be fast (he is just "sleepy", "lazy" when he plays or whatever), or his CNS is slow.

We can test that if he agrees to film some "glute unrelated movements" dynamically (can't come up with an idea now) or that keyboard type test (where you press the spacebar as fast as you can for 10 seconds and see what kind of numbers you come with to test "quickness" or CNS speed). I know, silly tests, but they might provide some information (relevant or not).

Back to the glutes idea, I would do dynamic glute movements if I were maxent - KB swings (a ton of them, focusing on the "hip snap"), med ball tosses (focusing on the hip snap), double leg bounds, depth jumps for length etc. I know a lot of these things have a horizontal component in them in terms of hip action - if anybody has any idea about a vertical hip dynamic movement... I'm all ears. Maybe SVJs to rim with little knee travel and a lot of hip bend/hinge (like what Toddday was trying to do in his depth jumps videos).

Plus sprints. If I were as slow as maxent, I would say "fuck this, I will do sprints like crazy" and try different distances and give 110% effort on them. Not only are they great plyometric exercises but also train the hip hyperextension, the posterior chain dynamically, the calves dynamically, stiffen the Achilles etc, all the good stuff.

Obviously, all these besides playing actual aggressive ball (especially on defense, since you don't have the ball and can move freely, you can be quicker/faster defensively and you'll benefit from that type of multi-directional "plyo work").

These are what I feel are the most sensible things for maxent at this moment.