Speed Cards Training

44 posts / 0 new
Last post
2 February, 2014 - 02:37
Offline
Joined: 4 years 11 months ago

Sheesh, I'm away from the forum for a couple of days and someone's editing my posts so it looks like I'm talking to myself... :)

But there's many a true word spoken in jest - my boss is the wildest party animal you could ever hope to meet; he sees any excuse for a night out (a group of Germans visiting from head office) as a prime opportunity to get everyone as drunk as possible. And, since this was a Thursday night, he of course expects his workers to be back in the office and none the worse for wear the first thing on Friday morning...

So since I was still feeling a bit woozy on Friday, and then more non-optional social obligations (I've been watching The Big Bang Theory a lot lately, and I really identify with Sheldon's definition of friendship :) ) forced me to have a drink at Saturday lunchtime and watch the football, I've returned to card-memorising and mnemotechnics.org-browsing today after three days off.

And it's made a noticeable difference for the better! I was significantly faster in my three attempts at speed cards today, and the recall wasn't as difficult as I've found it before. With the first pack I stopped the clock at 24.02 seconds, and just had to take a fifty-fifty guess at one blank image (one of the final five, I'd 'read' it as some gibberish, and nothing like what it should have been), and ended up putting the two cards together in the wrong order. The second time I did 23.93, and thought the recall was perfect, but I'd transposed the order of the final two images this time. Those final five don't normally give me problems, so it's interesting that I had two mistakes in a row.

In one of those funny coincidences, the first image of the second pack was the same image I couldn't remember in the first - what are the odds of that? But it didn't distract me and slow me down a little like funny coincidences usually do, so I was obviously really in the zone this morning.

The third pack I went a bit slower, and memorised it in 25.93 seconds, with a perfect and trouble-free recall.

It's making me wonder what is the cause of this improvement in speed? I can think of two possibilities - firstly that I've had that three-day break without memorising anything. I do always take a break like that before a competition, just so my journeys are fresh and I won't be confused by remembering images I've seen in practice, so maybe it's done me good here.

Or secondly, it's Sunday morning and the sun's shining for once. I've done most of my January training in the dark and rainy evenings after a day at the office, so maybe this is more of an optimum time for memorising? On the other hand, speed cards in competitions is always the last thing, after a long and arduous workload...

2 February, 2014 - 15:53
Offline
Joined: 4 years 3 months ago

Zoomy wrote:

It's making me wonder what is the cause of this improvement in speed? I can think of two possibilities - firstly that I've had that three-day break without memorising anything. I do always take a break like that before a competition, just so my journeys are fresh and I won't be confused by remembering images I've seen in practice, so maybe it's done me good here.

Systematic recording of the amount of time spent practicing a particular skill, the break time between practice sessions, the practicing of similar skills in the meantime (such as digits one day and random words the next day), and the rate of improvement would not only tap practical knowledge that could be used to schedule training, but would also be a boon to cognitive neuroscience in theory because it could provide more clues about what it means for one skill to be "like" another, and on what levels the similarity is relevant. Questions like "does interference slow progress when a person works on digits one day and then binary the next, because they use the same images? Or Digits one day and random words the next because they both use loci? The same loci? How about beginning to learn an image list one day and French the next? I haven't been able to find very detailed questions to these answers, or where the people trying to solve them do their research.

That's a bit off topic, though. I always improve the most during the 24 or more hours that I don't work on a skill, even (especially!) if I've been doing it for 4 or more hours each day in a row up until that break.

Cards today:
1)40.33. Barely had enough time, but I got the second deck in order! Skipped the "safe run" because I haven't been failing recently anyway.

Here's something bizarre:
The Big Bang Theory with the laugh track removed.

http://www.youtube.com/watch?v=jKS3MGriZc

Clips of sitcoms with the laugh track removed are not funny, and that fills me with self-loathing.

Edit:
Feb 3: My first deck today took 0:50.81; failed to reconstruct. When I drink beer, my memory and general cognitive speed takes a very significant hit for about 36 hours, so a drop in time like that is not out of the ordinary after a few beers the night before (Superbowl party :\). I DID encode an wrong image for no good reason, just saw the cards and thought of something different. That's scary stuff, but when it happens I just review ALL of my images and it won't happen again for another month. But doing that is a boring way to spend an hour -- how on earth can you do that in 15 minutes?! If you could memorize at that speed you'd be knocking out a deck in under ten seconds!

Edit: Feb 4,
Reviewed through all of my face-cards combos' images twice. That alone was more than enough for one day. After doing that the images really pop out, so I put the metronome up to a quick pace and gave it a couple tries.
1) 0:31.22. Fail. Couldn't remember 20 cards so I didn't bother messing up my ordered deck to try to reconstruct it.
2) 34.88. This one was just as bad :) I'll do one more without a metronome so I can control my own pace, even though I should just be practicing random words instead :/

3) 0:52.76. Took a 50/50 guess on one image and got it right.

4 February, 2014 - 11:49
Offline
Joined: 4 years 11 months ago

After a night off from the cards yesterday (either to test the theory of improved times with a break between training sessions, or just because I've been working extra hours at the office and just couldn't be bothered), tonight was a good one - 26.58, correctly recalled although it was a struggle; picked up the wrong card a couple of times and had to correct myself, so in the end the recall took nearly the whole five minutes, then 25.86 with a mistake in the 'blind spot' (the last couple of images before the final five), and finished with 24.88, recalled correctly without much trouble.

The ideal is to be able to do 25 seconds consistently enough that I can bank on it as a 'safe' time, and then go for the world record in the second trial, but we're still a long way away from that...

Hope your team won the Superbowl, anyone who's reading this! Chelsea beat Man City, so I'm happy, more or less. :)

5 February, 2014 - 11:39
Offline
Joined: 4 years 3 months ago

1) 58.71. Reconstructed fast, but forgot 5 images. It's been a long time since memorizing a deck even slowly without forgetting images and I'm started to think I've developed the habit of "good enough" memorizing -- which isn't always good enough.

2)0:41.82. Barely, barely put it back. I've been just over 0:40 at my own pace recently, so I'm gong to do a couple with a metronome and bring it down a few seconds.
3) Gaw!! That didn't go well. Didn't sleep either of the last two nights and I'm beginning to think that is contributing to the overall poor performance today, even though it usually doesn't.

6 February, 2014 - 13:14
Offline
Joined: 5 years 2 months ago

I'm now dipping in the 3 minutes range with my morning packs and in the 4 minutes range with my after dinner packs.
This came with a little surprize. I achieved a 3:17 with 4 errors on a morning when I had very important business to attend. Never before had I dipped below 4 minutes with the method that I used. In my mental preparation, I kept repeating to myself that I wanted to go so fast as to make about 15 errors just before starting.
Since then, I haven't gotten close to that score; my scores have just kept climbing so far this week but I've been able to keep my times below 4 minutes. I guess it shouldn't be too long now before I dip below 3 minutes, with 15 errors.

Simon

8 February, 2014 - 19:09
Offline
Joined: 4 years 3 months ago

The 2 runs today:
1) safe run: 0:57.07 - Failsters. Drank a bit last night and am unfocused today... Half way into the deck I realized that I had forgotten that this was a safe run and was automatically moving at a quick "speedrun pace," and so shifted into an absurdly sluggish second half to try to make up for it. Which was dumb - I was comfortable before I did that, so I should have just let it be. Anyway, misplaced 2 adjacent images during reconstruction. So now I have to do a second run that is safe-as-a-straightjacket, just like I would in the real deal. So that will probably be like a 1:00 safe-run and then full review afterward.

2) 1:45.69. Success. Mem'd deck in 1:15 and then reviewed the deck in 0:30. Even at that fast review pace I predicted over 40 of the cards before I saw them, so the review had to be unnecessary, but...I guess that's the point. Days like this make me really nervous about the strategies to employ at the USAMC.

Ben, you must agree...that even a safe time is not totally safe? But you don't let this scare you into going absurdly slow or reviewing the pack, right? Has that ever screwed you up in a competition, not being safe enough?

Update, Feb 12:

1) safe run: 1:00.28, perfect recall, first time in a while!
The days after I go through and review all of my images with Josh's Ben System Worksheet, the cards and images are so natural. Over the past few days, I've screwed up safe runs with digits and cards so many times that I've had to reconsider what kind of handling is "safe." So I took the time to really link everything up nicely and see it really clearly, to try to remember it all, and the safe time was the same as always! That's what a full review does.

2) Speed run: 0:42.48. Fail. During reconstruction, put the wrong cards down for one of the earlier images...remembered the image, but wrong cards. Which means that I didn't have the right cards left over in my hand. So put together 48 cards and then was just confused about the last 4. If I hadn't made that mistake, I think I could have gotten it back together...but it's the kind of error that I don't know how to forestall for the future. Cognitive lapses 'just happen' from time to time.

12 February, 2014 - 09:14
Offline
Joined: 5 years 2 months ago

LociInTheSky wrote:

Your method for checking through the remaining cards as I understand it is kind of unexpected. It's inefficient. Let's say it was a deck where you did have 16 cards worth of blanks. If you were to check 15 images with one card and then put it at the back if it doesn't produce one of the images then you could put together 134 incorrect images before coming up with one of the ones that was actually in the deck. If you checked the combinations where a given card was first, and none of them checked out, and then you checked the combinations where that card was second, you could only put together a maximum of 29 incorrect images before coming up with one that was in the deck. If you can put a deck together in only 2:00, then changing the way you handle the remaining cards might save enough time to put the decks back together more often. Don't you think?

I'm surprised to see Zoomy agreeing with you on this one. Overall, there is no gain in efficiency, in my opinion. If I understand the process correctly, the only gain in efficiency that your method suggests is in finding a first image quickly.
My understanding is that the same amount of images will have to be checked any way. It's a bit like having to check for a bean in a cake and Zoomy checks only half and then he comes back around. You seem to suggest looking at the whole cake first. Given that there is a bean in every cake you will find the image sooner but overall, it is the same thing. Also, given the way that Zoomy has memorized his images on sheets of 52 images per pages all starting with the same card, then it would indeed make more sense to keep zooming through this mental page than flipping the mental pages over to find the connection from the other way around.
According to my calculations, the chances of putting together 134 images together without coming up with one are very small indeed.

Simon L.

12 February, 2014 - 13:11
Offline
Joined: 4 years 3 months ago

simon L. wrote:

Overall, there is no gain in efficiency, in my opinion.

According to my calculations, the chances of putting together 134 images together without coming up with one are very small indeed.

Zoomy can not help but agree because the increase in efficiency is not a matter of opinion. For any amount of cards that are left over, by the time you would be guaranteed to find your first image using the method I suggested, there would be a 25% chance that you would not have found the image at all using Zoomy's method, and furthermore, up until that point, the odds of any "next pair" being an image that you are looking for increases by a greater margin using my method. And though it is true that the odds are very small of the card-pairs you are looking for being the absolute last ones you check, running out of time on account of not having found the right cards before time is up would occur in many possible situations other than that worst scenario. So one out of every 32 decks Zoomy memorizes, he will sift through nearly four times as many images as the maximum that I would sift through before finding the first image that he is looking for. "1 deck out of 32," you may think, "that's so unlikely." It depends how you look at it. If you only enter one competition in your life, then yes, it is unlikely. But if you continue to compete, the law of large numbers sets in, and the odds of sabotage approach certainty. Ben has already competed in 39 championships, memorizing 2 decks for most or all of those, which means that the odds are extremely good that this has already happened to him. Keep in mind that finding the first image is very important because after it is found, there are fewer cards left in the bank to make images from, so the next images will probably be found sooner. So after you have found your first images, each image you look at after that yields a greater probability of being a sought image than those images sifted through by the counterpart who has not found his first correct first image, so that the disparity of efficiency actually widens over time.

If you prefer analogies, here is an apt one:

Say you have 15 bags, each with 2 white balls and 1 black ball. Being able to get a black ball quickly may be vital to your success at this game. So you reach into the first bag, and pull out a white ball.

Too bad. But you get to go again. So which bag do you reach into now? I would reach into the first bag again, because the elimination of the first ball means that there is a 50% shot you'll get the black ball on the next draw, and a 100% that you'll get it within the next two draws. We can talk all we want about how it isn't likely that by drawing balls all around at random, you'll pull only the white balls out of all the bags and leave the black ones in, but you wear a seat belt when you drive, don't you? There is nothing to gain for reaching into a different bag on your second try, where all those bags give you 33% odds versus bag 1 which gives you 50% odds. And you see how the probability increases over time because let us say that on your third try you reach again for a new bag, keeping your chances at 33% while mine have moved all the way to 100%, and remember that after I have the first ball, all of my draws from then on yield a higher probability of getting a black ball (not true in this "bag" analogy, but true with cards). So as I move on, you might be left in the dust. Then again, you might not. But either way, I'm looking good.

Zoomy may have decided that it is not worth the effort to make the switch but I humbly suggest he reconsider. This is not something that he cannot learn to do perfectly well with a little practice, and that's time well spent because Russian Roulette with a 64-shooter is still Russian Roulette.

Edit: Just memorized another deck and it was a PB! My fastest time memorizing a deck and being able to recall the order of the deck instantly = 0:41.90. Now I'm going to fight through the pain and review all of the images again, it's too helpful not to.

13 February, 2014 - 09:28
Offline
Joined: 5 years 2 months ago

LociInTheSky wrote:
simon L. wrote:

Overall, there is no gain in efficiency, in my opinion.

According to my calculations, the chances of putting together 134 images together without coming up with one are very small indeed.

Zoomy can not help but agree because the increase in efficiency is not a matter of opinion. For any amount of cards that are left over, by the time you would be guaranteed to find your first image using the method I suggested, there would be a 25% chance that you would not have found the image at all using Zoomy's method, and furthermore, up until that point, the odds of any "next pair" being an image that you are looking for increases by a greater margin using my method. And though it is true that the odds are very small of the card-pairs you are looking for being the absolute last ones you check, running out of time on account of not having found the right cards before time is up would occur in many possible situations other than that worst scenario. So one out of every 32 decks Zoomy memorizes, he will sift through nearly four times as many images as the maximum that I would sift through before finding the first image that he is looking for. "1 deck out of 32," you may think, "that's so unlikely." It depends how you look at it. If you only enter one competition in your life, then yes, it is unlikely. But if you continue to compete, the law of large numbers sets in, and the odds of sabotage approach certainty. Ben has already competed in 39 championships, memorizing 2 decks for most or all of those, which means that the odds are extremely good that this has already happened to him. Keep in mind that finding the first image is very important because after it is found, there are fewer cards left in the bank to make images from, so the next images will probably be found sooner. So after you have found your first images, each image you look at after that yields a greater probability of being a sought image than those images sifted through by the counterpart who has not found his first correct first image, so that the disparity of efficiency actually widens over time.

If you prefer analogies, here is an apt one:

Say you have 15 bags, each with 2 white balls and 1 black ball. Being able to get a black ball quickly may be vital to your success at this game. So you reach into the first bag, and pull out a white ball.

Too bad. But you get to go again. So which bag do you reach into now? I would reach into the first bag again, because the elimination of the first ball means that there is a 50% shot you'll get the black ball on the next draw, and a 100% that you'll get it within the next two draws. We can talk all we want about how it isn't likely that by drawing balls all around at random, you'll pull only the white balls out of all the bags and leave the black ones in, but you wear a seat belt when you drive, don't you? There is nothing to gain for reaching into a different bag on your second try, where all those bags give you 33% odds versus bag 1 which gives you 50% odds. And you see how the probability increases over time because let us say that on your third try you reach again for a new bag, keeping your chances at 33% while mine have moved all the way to 100%, and remember that after I have the first ball, all of my draws from then on yield a higher probability of getting a black ball (not true in this "bag" analogy, but true with cards). So as I move on, you might be left in the dust. Then again, you might not. But either way, I'm looking good.

Zoomy may have decided that it is not worth the effort to make the switch but I humbly suggest he reconsider. This is not something that he cannot learn to do perfectly well with a little practice, and that's time well spent because Russian Roulette with a 64-shooter is still Russian Roulette.

Edit: Just memorized another deck and it was a PB! My fastest time memorizing a deck and being able to recall the order of the deck instantly = 0:41.90. Now I'm going to fight through the pain and review all of the images again, it's too helpful not to.

You're absolutely right. I withdraw my earlier comment. Thanks for making things clear to me with this analogy.

Simon

14 February, 2014 - 05:52
Offline
Joined: 4 years 3 months ago

:) You're very welcome.

Edit, Feb 17:
Today I've taken the first step to lower my card times down to the next level. Covered 1/8th of that journey pretty thoroughly over the past hour or so. Since cards are the final event at the USAMC, I really feel that I should be more prepared in case I get that far. Of the images that take the most time to process, probably 90% are shadow images. With a little work, those are going to be the faster half because of their abbreviated names. I already say the name when I see the cards, but more processing is usually required to get to the image because the syllable doesn't have the same amount of meaning to me that an English word does. A long time ago I made all of the flashcards for this purpose, but it's a lot of work, and I put it off for a while because I was improving without it. So I'm going through right now and making my 1352 mnemonic connections between the syllable and the image. Over time, the strength of this connection will overpower the strength of each shadow abbreviation's association with its respective original image pair (sorry if I lost some of you there). That will lop off a giant chunk of time. I think that I can become fluent with all of those sounds by late March, and I'll keep everyone apprised.

26 March, 2014 - 16:43
Offline
Joined: 4 years 3 months ago

Just finished up my last four decks before the break before USAMC.
1)1:01.34 perfect
2) 43.75 perfect,

Both consistent with my nearly metronome-perfect speed and safe runs over the past few months...crazy. Then

3) Really quick run - 34.45 Reconstructed with no difficulty.

Good for my confidence! Leaving at 5:00 AM tomorrow so I'm looking forward to seeing/meeting you guys in NY, later then!

26 March, 2014 - 17:01
Offline
Joined: 5 years 1 month ago

Very impressive, indeed!

20 March, 2017 - 07:06
Offline
Joined: 1 week 6 days ago

Hi!

Newbie here.

This thread was really inspiring. I haven't tried to memorize the whole deck yet (I've been toying with memorization stuff, ironing out my coding systems with a number of weeks now). I started with just 14 cards, then 16, and last night, 18. I'll add two cards every day/every other day until I have a full deck. Then I'll turn on a stop watch and start timing myself.

Also, in the meantime, I'm trying to solidify my journeys for cards. So, I'll probably do 20 cards tonight with a different journey than yesterday, 22 the next day with a different journey, etc. I have 10-15 journeys in mind, so I'll spend my commute time really visualizing the loci within those journeys.

Would love to hear about other people's training some more :)

Learn memory techniques for free! Just click the "Sign up" button below to create an account and we'll send you an email with some tips on how to get started.

Related content: