Older cubers discussions

[SpartanSailor]

I filmed my attempts for the weekly competition but no success but I was eager to get my first filmed success so did a couple more solves and in the end got this on film:-

1. 9:46.02 F2 D' B2 R2 U2 F2 R2 D B2 D' U' L' F U2 B U' L' F' D' F2 U2

Sub UK hard cut but my aim is to get sub 5 minutes for High Wycombe Open in April to have a couple of official attempts.

Well done! I notice that you did Corners First... is this how you typically do it? I get my corner memo first, then do edges in execution first. If there is an odd number of targets, I deal with that prior to commencing my corner memo. I've thought to do corners first, but wonder if the "fix" for odd targets is results in the same outcome. Again, nicely done.

I'm not sure what the hard-cuts are in the UK, but here in the States along the EAST coast there are few comps (currently, at least) that do 3BLD and I've seen the hard-cut from 6 min to 10 min. Our US NATIONALS qualifying time is 4 minutes for 3BLD. I am in the 9ish minute range... Sometimes a little faster, and other times a little slower--my issues come when I'm returning my set-up moves. I sometimes do them the wrong way and the cube is no longer aligned for the rest of the solve... results in a very unimpressive attempt of a cube that looks fully scrambled despite my 8 minutes of memo.

 

[newtonbase]

I filmed my attempts for the weekly competition but no success but I was eager to get my first filmed success so did a couple more solves and in the end got this on film:-

1. 9:46.02 F2 D' B2 R2 U2 F2 R2 D B2 D' U' L' F U2 B U' L' F' D' F2 U2

Sub UK hard cut but my aim is to get sub 5 minutes for High Wycombe Open in April to have a couple of official attempts.

Sub 10 is excellent and I'm sure you'll get sub 5 by April.

Did my first proper 5BLD attempt and very happy with the result. IncirInco pieces were 5 plus cross (recall and/or memo), 1 flipped midge (memo) 2 twisted corners (comm error). Wings and X cross all good. Took just over 40 mins being super careful so I'm good for the Manchester time limit.

 

[Selkie]

@SpartanSailor - I memo edges first using images and then memo corners using audio. I then execute corners first from short term audio before moving on to edges. If both have odd number of targets then its parity after edges, if both even, no parity. One odd one even I sigh and look for my memo mistake

I think it is usual in UK to have a total of 10 minutes or 3 attempts whichever comes first so if I can get down to below 5 minutes I should be able to squeeze in more than one DNF

I have to say at this point in time I think audio memo is actually counterproductive to my level as I have to keep repeating the audio loop to myself given the amount of time it takes me to memo the corners but I wanted to give it a go given it will be beneficial later. I also get stuck trying to think of some short sounds for some pairs. I use "Sh" sound for X and "Ch" for Q but still need to differentiate between C and K. I feel an audio letter pair list on the horizon!

Talking of letter pair lists I have now completed the full images list but I am sure some will need changing. I wanted to do this as quickly as possible so I can start learning them as soon as possible. It is amazing what sticks as my UQ happens to be your dog headed slug @mark49152 , its the first thing that came to mind when I got it in memo. Though my IQ is a scientist with a head like in the Tefal adverts in the 80's I have used QI for monocle.

I used all of the very good letter pair resources kindly offered here last week but I also found this quite useful for some tricky ones .. https://www.acronymfinder.com/

Going to try and fit in 5 BLD solves a day through February and keeping a record of DNF proportion and overall times.

Looking forward to getting better at the lettering scheme and tracing and I am very eager to move on to M2 as soon as possible but overall happy with progress so far.

@newtonbase - That sounds awesome for your first 5 attempt, not much wrong with that mate.

 

[pglewis]

Well done on the sub 1 @Ander. It's a great milestone that i got to experience twice as the first time turned out to be a DNF.

Oh man, that's harsh

I've thought to do corners first, but wonder if the "fix" for odd targets is results in the same outcome.

I'm nearly 100% certain you need a different parity fix if you execute corners first (but I've always done edges first).

 

[newtonbase]

I have to say at this point in time I think audio memo is actually counterproductive to my level as I have to keep repeating the audio loop to myself given the amount of time it takes me to memo the corners but I wanted to give it a go given it will be beneficial later. I also get stuck trying to think of some short sounds for some pairs. I use "Sh" sound for X and "Ch" for Q but still need to differentiate between C and K. I feel an audio letter pair list on the horizon!

I use Sh for X, Th for Q and Ch for C. Where H is the 2nd letter I'll usually use Ng. DeeDubb did a thread on audio pairs which really helped. https://www.speedsolving.com/forum/...ocus-on-phonetics-rather-than-spelling.51937/

 

[SpartanSailor]

So letter pair stuff... I know you guys are talking about images for each letter pair. Having only 1 image per pair, makes a decent listing... (3BLD...can't imagine 4, 5 and other big blind). I haven't taken to the visual method, but should probably give it some attention. I just brute-force my way through all the letters and try to string some non-sense sentence together.

 

[mark49152]

It is amazing what sticks as my UQ happens to be your dog headed slug @mark49152 , its the first thing that came to mind when I got it in memo. Though my IQ is a scientist with a head like in the Tefal adverts in the 80's I have used QI for monocle.

Haha great, you are welcome . My IQ is the crazy guy out of Back to the Future. I also have EI = Einstein. For QI, I use quiet which is on my "bad, must change" list. Sometimes I visualise it as a zip on someone's mouth.

 

[pglewis]

So letter pair stuff... I know you guys are talking about images for each letter pair. Having only 1 image per pair, makes a decent listing... (3BLD...can't imagine 4, 5 and other big blind). I haven't taken to the visual method, but should probably give it some attention. I just brute-force my way through all the letters and try to string some non-sense sentence together.

I'm approaching memo and letter pair images primarily with mbld in mind: I'm mostly concerned about a system that can stick for me for multiple cubes. That approach will be slower for 3bld but it certainly won't be a bottleneck to reaching sub 5 so it's good enough for both 3bld and mbld for the short term.

Images work fairly well for me but everyone is different. If I tell a story I tend to visualize that story somewhat and the visual memory for even a short, simple story encodes a surprising amount of information. The amount of information for a single 3x3 using an image per letter pair is surprisingly small. I mentioned "Albert Einstein on Roller Skates" as an example of "AE RS" before and you said you had difficulty visualizing, that it was just words. It might help to bring some action into the image to help really turn it into an image. How is Al doing on those skates? Maybe he's surprisingly good, landing triple axles. Or maybe he's struggling, legs akimbo, desperate not to fall. I'll often use tricks like this to link image to image so I can follow it like links in a chain and the main trick is usually remembering that first corner or edge image, if I have that first link I'm usually good. My image examples might not be good for you and some slightly different approach might be better suited, too. Note that some of Mark's memorable images don't even have a language connection to the letter pair but they work just as well once embedded.

It is definitely a lot of images to collect and I'm playing that as a long-game.

 

[newtonbase]

Images work fairly well for me but everyone is different. If I tell a story I tend to visualize that story somewhat and the visual memory for even a short, simple story encodes a surprising amount of information. The amount of information for a single 3x3 using an image per letter pair is surprisingly small. I mentioned "Albert Einstein on Roller Skates" as an example of "AE RS" before and you said you had difficulty visualizing, that it was just words. It might help to bring some action into the image to help really turn it into an image. How is Al doing on those skates? Maybe he's surprisingly good, landing triple axles. Or maybe he's struggling, legs akimbo, desperate not to fall. I'll often use tricks like this to link image to image so I can follow it like links in a chain and the main trick is usually remembering that first corner or edge image, if I have that first link I'm usually good. My image examples might not be good for you and some slightly different approach might be better suited, too. Note that some of Mark's memorable images don't even have a language connection to the letter pair but they work just as well once embedded.

I do this sort of thing too, especially on a weaker image. Focusing on a minor detail really helps it stick. I saw the technique suggested on a video. Ollie's maybe?

 

[Mike Hughey]

Images work fairly well for me but everyone is different.

This is very true! Some people take very well to visualizing; others do not. It doesn't seem to matter much for success, though - you need to use what's best for you.

I find that I do much better when I can really see the visuals while memorizing. But sometimes when I'm trying to go too quick, I just get raw words, and when that happens, I'm much more likely to forget later on in recall.

But one thing that is very true is that visuals seem to take longer. It takes me precious seconds to really implant a visual image in my memory, and it seems like the fastest people just don't do that. Mark Boyanowski (world #3 in multiBLD with 40/42 in competition) told me that he felt like he was handicapped for BLD because he couldn't do mental images well at all, but then he talked a bit with Marcin Kowalczyk (world #1) and Marcin told him that he mostly doesn't even use images - he just does sentences that are basically just words. He said they were faster. And it seems like Mark has had a lot of luck with that.

So, my point is, if you're the type who works better with sentences that are just words, rather than images, and it actually works for you, maybe that's to your advantage!

It certainly seems to me that to get the sub-20 3x3x3 BLD times that the fastest people get, there must be absolutely no imaging going on at all - just audio loop of the entire memorization and go. With just 7 seconds to memorize a cube, there really isn't time for more than that. It's just so hard to comprehend how they do that. When I switched from audio for corners (with images for edges) to audio for edges (with images for corners), I actually slowed down substantially, and I have yet to recover. I haven't had time to really drill it; I keep hoping that if I do, I'll eventually get faster and it will pay off. But it hasn't happened yet.

I do this sort of thing too, especially on a weaker image. Focusing on a minor detail really helps it stick. I saw the technique suggested on a video. Ollie's maybe?

One other trick that can help is to make an image from one location in your route interact somehow with an image from a nearby location. I find that can really help, especially with some image that is otherwise not sticking very well.

An example of this from my most recent multi attempt: I had Johnny Depp using a guage to measure a cat's hiccups. My next location was of a quivering sail on a sailboat. I couldn't get it to stick, then I realized I could have Johnny Depp (as Captain Jack Sparrow) reaching over from the other location and messing with the sail, and then it stuck so well that I still remember it.

 

[SpartanSailor]

I'm still in my infancy as a BLD solver... having only just started a few weeks ago, I am way too novice to speak with certainty about what "works" for me and what doesn't. Just for now, that is how my brain has begun processing things. I will probably spend some time working on images. I DO like the idea that images are connected to "pairs". I spend MOST of my time trying to determine if I've hit every target then trying to count how many targets there were... Just a string of words doesn't make it intuitive whether there are even or odd targets. I see the advantage that each image is an even number of targets. So.... if there's a left over target, then I add the parity alg before starting corners. More simple in my mind that way.

Plus, I like the image of Albert Einstein doing a triple axle and trying not to fall on ice skates... I have to work that in somehow!

 

[CLL Smooth]

Blindfold still?!?! Sheesh, you guys kill me. Don’t you talk about anything else in here?
So I’m doing blindfold now and my brain actually hurts. Perhaps if I stopped smoking the reefer my short term memory would improve.

 

[pglewis]

I'm still in my infancy as a BLD solver... having only just started a few weeks ago, I am way too novice to speak with certainty about what "works" for me and what doesn't. Just for now, that is how my brain has begun processing things. I will probably spend some time working on images. I DO like the idea that images are connected to "pairs". I spend MOST of my time trying to determine if I've hit every target then trying to count how many targets there were... Just a string of words doesn't make it intuitive whether there are even or odd targets. I see the advantage that each image is an even number of targets. So.... if there's a left over target, then I add the parity alg before starting corners. More simple in my mind that way.

Plus, I like the image of Albert Einstein doing a triple axle and trying not to fall on ice skates... I have to work that in somehow!

Experiment, use your creativity and see if you can find what works well for you. I'm working to get tracing more automatic so I can confidently get through a full 3bld trace reasonably smoothly without review. No use fighting both tracing and assigning images at the same time, if tracing is as quick as it should be then I won't even have to retain the memo very long for just 3bld, execution doesn't take long.

 

[SpartanSailor]

The discussion DOES tend to center around various blind solving events (3BLD, bigger cubes, multi)...

But, since you asked.... I'm holding my nerves at bay. I have a comp this weekend in Pennsylvania. I'm competing in 2x2, 3x3, 4x4 and skewb. I haven't given the skewb any real attention lately, which is too bad. Sometimes, I'll have an amazing average and blow right by my official PB, but then my consistency is out the window... so that could go either way. Same for 2x2.

I am hopeful that I can deliver with 3x3. I'm not the fastest, by any stretch, but I have made a relatively decent breakthrough somewhat recently. I had a comp last month and actually did pretty well, even with a +2 on my fastest I still got a PB single and PB Ao5. Since then, I have really seen a dramatic improvement. But, I get serious competition nerves and usually have 20s and 21s while I'm warming up or just messing around, then blow it with high 20s and sometimes 30s in comp. Sooooo.... I'll be thrilled if I can deliver on my 3x3 and just avoid embarrassment with 2x2 and skewb.

 

[SpartanSailor]

Experiment, use your creativity and see if you can find what works well for you. I'm working to get tracing more automatic so I can confidently get through a full 3bld trace reasonably smoothly without review. No use fighting both tracing and assigning images at the same time, if tracing is as quick as it should be then I won't even have to retain the memo very long for just 3bld, execution doesn't take long.

Exactly. If I was confident I had every target, I wouldn't need much to remember my memo... I spend so much time trying to trace that I end up having to recreate my memo a couple times.

 

[newtonbase]

This is very true! Some people take very well to visualizing; others do not. It doesn't seem to matter much for success, though - you need to use what's best for you.

I find that I do much better when I can really see the visuals while memorizing. But sometimes when I'm trying to go too quick, I just get raw words, and when that happens, I'm much more likely to forget later on in recall.

But one thing that is very true is that visuals seem to take longer. It takes me precious seconds to really implant a visual image in my memory, and it seems like the fastest people just don't do that. Mark Boyanowski (world #3 in multiBLD with 40/42 in competition) told me that he felt like he was handicapped for BLD because he couldn't do mental images well at all, but then he talked a bit with Marcin Kowalczyk (world #1) and Marcin told him that he mostly doesn't even use images - he just does sentences that are basically just words. He said they were faster. And it seems like Mark has had a lot of luck with that.

So, my point is, if you're the type who works better with sentences that are just words, rather than images, and it actually works for you, maybe that's to your advantage!

It certainly seems to me that to get the sub-20 3x3x3 BLD times that the fastest people get, there must be absolutely no imaging going on at all - just audio loop of the entire memorization and go. With just 7 seconds to memorize a cube, there really isn't time for more than that. It's just so hard to comprehend how they do that. When I switched from audio for corners (with images for edges) to audio for edges (with images for corners), I actually slowed down substantially, and I have yet to recover. I haven't had time to really drill it; I keep hoping that if I do, I'll eventually get faster and it will pay off. But it hasn't happened yet.


One other trick that can help is to make an image from one location in your route interact somehow with an image from a nearby location. I find that can really help, especially with some image that is otherwise not sticking very well.

An example of this from my most recent multi attempt: I had Johnny Depp using a guage to measure a cat's hiccups. My next location was of a quivering sail on a sailboat. I couldn't get it to stick, then I realized I could have Johnny Depp (as Captain Jack Sparrow) reaching over from the other location and messing with the sail, and then it stuck so well that I still remember it.

I like to play around with them too. I don't have fixed numbers of pairs in one image. Today I had Mo puts a QuiFf in a shed (XD). The next 2 were TR AW, usually Tear Amy Winehouse which didn't work so I put TyRes in the shed and started again with Amy (and her He-Pig).

 

[Jason Green]

This is very true! Some people take very well to visualizing; others do not. It doesn't seem to matter much for success, though - you need to use what's best for you.

I find that I do much better when I can really see the visuals while memorizing. But sometimes when I'm trying to go too quick, I just get raw words, and when that happens, I'm much more likely to forget later on in recall.

But one thing that is very true is that visuals seem to take longer. It takes me precious seconds to really implant a visual image in my memory, and it seems like the fastest people just don't do that. Mark Boyanowski (world #3 in multiBLD with 40/42 in competition) told me that he felt like he was handicapped for BLD because he couldn't do mental images well at all, but then he talked a bit with Marcin Kowalczyk (world #1) and Marcin told him that he mostly doesn't even use images - he just does sentences that are basically just words. He said they were faster. And it seems like Mark has had a lot of luck with that.

So, my point is, if you're the type who works better with sentences that are just words, rather than images, and it actually works for you, maybe that's to your advantage!

It certainly seems to me that to get the sub-20 3x3x3 BLD times that the fastest people get, there must be absolutely no imaging going on at all - just audio loop of the entire memorization and go. With just 7 seconds to memorize a cube, there really isn't time for more than that. It's just so hard to comprehend how they do that. When I switched from audio for corners (with images for edges) to audio for edges (with images for corners), I actually slowed down substantially, and I have yet to recover. I haven't had time to really drill it; I keep hoping that if I do, I'll eventually get faster and it will pay off. But it hasn't happened yet.


One other trick that can help is to make an image from one location in your route interact somehow with an image from a nearby location. I find that can really help, especially with some image that is otherwise not sticking very well.

An example of this from my most recent multi attempt: I had Johnny Depp using a guage to measure a cat's hiccups. My next location was of a quivering sail on a sailboat. I couldn't get it to stick, then I realized I could have Johnny Depp (as Captain Jack Sparrow) reaching over from the other location and messing with the sail, and then it stuck so well that I still remember it.

That's good to hear because I have a hard time doing images as compared to sentences. If I can have a decent sentence I can visualize the whole thing sometimes, but it's more the story of what's going on than an image.

 

[pglewis]

Blindfold still?!?! Sheesh, you guys kill me. Don’t you talk about anything else in here?

It's the one place we can still stay ahead of most of the youngsters. For now. They're quickly discovering it though, those days are numbered

 

[pglewis]

One other thing, if it's just 3bld then I don't really have to stick an image to everything. I just got an untimed success where I didn't even bother coming up with images for the last few pairs. Corner memo has to stick longer, I just had 5 images. I had 4 images for edges and just remembered "BT CW S(parity)" for the end, trusted my trace and did edges before I could forget. It's not unusual for me to remember some portions of my trace just spatially... if I could do that 100%, who even needs Speffz, let alone images?!

 

[Mike Hughey]

It's not unusual for me to remember some portions of my trace just spatially... if I could do that 100%, who even needs Speffz, let alone images?!

There was a time when that was the "in thing" - that's what was meant by "visual memory". People described it as remembering an arrowed path that runs from piece to piece, as if it's a 3d map that you're remembering for the memo. But since then it seems like the people who just glom a bunch of letters together in some pronounceable way for an audio loop
have become the fastest ones.

It's the one place we can still stay ahead of most of the youngsters. For now. They're quickly discovering it though, those days are numbered

Tell me about it. There was a time not very long ago when 12/12 multi would have easily been US national champion material. This year it was barely enough for 7th place.