Message boards :
News :
Onwards to new and great (actually, lesser) heights (of cacti)
Message board moderation
Author | Message |
---|---|
Send message Joined: 14 Jun 20 Posts: 78 Credit: 1,321,619 RAC: 0 |
The first round of Kaktwoos processing is nearing its completion. A small number of hosts caused many errors with faulty hardware or software configurations which caused the failure of a handful of workunits. Due to these failures; several thousand new tasks have now been submitted to allow any previously failed workunits an additional chance to succeed. All historically submitted data is currently under re-evaluation as a subset of these completed tasks has outputs which don't appear to be nominal. The re-validation of tasks will reach completion shortly, and soon a follow-up will be posted detailing the tallest cactus found during this round of Kaktwoos processing. The Kaktwoos (kak-two-s) project was initiated by computing a small set of tall 'candidate cacti' using the single-chunk cacti stacking application Kaktoos (kak-tus). The initial cactus seed search assumed stacking began at an elevation of y=63 blocks from bedrock which is the basis of assumption for the Kaktwoos multi-chunk cacti scanning work. You have likely volunteered resources towards this effort over the past few weeks. Now, onto the exciting news: the Kaktoos application (note the missing 'w') shall be launching on BOINC. This application will exist to scan for individual chunk seeds which produce tall cacti (7 blocks minimum) with generation elevation of y=62. Using a lower elevation parameter for Kaktoos will hopefully allow us to discover additional naturally-generated 22+ block tall cacti with Kaktwoos. For those wondering, the next set of tasks will produce many more resultant 'candidate' cactus seeds than the previous y=63 run. We expect 120 million generated seeds despite the required minimum height of 7 blocks we chose to reduce output drastically. This dataset will be used for quad-chunk stacking (Kaktfours anyone?) and is much more challenging to manage due to the lower probability of finding 22+ block tall cacti and increased development overhead. We could theoretically find much taller cacti with this method, with a caveat of requiring significantly more processing power. Don't worry though; there are other projects brewing which will take priority. If you are a 'stacktus' superfan or a member of the 'big cactus community', you also needn't worry; there may be even more permutations of Kaktoos and Kaktwoos inputs to try which should keep us all busy along the way. Finally, and for reference, the tallest known cactus found by MC@Home (in Minecraft: Java Edition v1.14.4) is 22 blocks, and 24 blocks tall if using a superflat world with the desert preset. We distinguish these heights as we consider "cactus naturalis" (cactus of normal generation) cacti to be the true tall cacti, as opposed to "cactus fma vi" (cactus of superflat generation). |
Send message Joined: 25 Jul 20 Posts: 1 Credit: 3,151,117 RAC: 0 |
looking forward to it! |
Send message Joined: 15 Jun 20 Posts: 74 Credit: 19,537,761 RAC: 0 |
Just a bump to this thread, as we are preparing to begin the new set of tasks. All Kaktoos (prelim part-2 cactus search) tasks are essentially done, meaning we have a lot more cactus to attempt stacking for 22+ or taller cacti. We found some very nice 13 tall cactus that might make it easier to go higher than 22 tall as well. |
Send message Joined: 26 Jun 20 Posts: 25 Credit: 123,735,290 RAC: 66 |
We're waiting :) |
Send message Joined: 15 Jun 20 Posts: 74 Credit: 19,537,761 RAC: 0 |
The waiting game is over, as we have begun Kaktwoos 'y=62' (Round 2) of cactus searching. We are attempting to stack cacti together on sand that is 1 block lower than before, which may potentially provide another set of fresh, 22+ tall cacti to find. Wish everyone luck in their stacking and MC@Home's cactus search is officially re-opened. |
Send message Joined: 15 Jun 20 Posts: 74 Credit: 19,537,761 RAC: 0 |
Small update! We're moving to finding Pack.PNG ('PackCrack CUDA / Nvidia) on BOINC. At the moment we're very close to finishing this current run of Kaktwoos, which is amazingly fast. We shall be looking through all the images, but For the first y=62, 22 tall! A beachside cactus Credit to Dutchen18 on our Minecraft@Home discord, who has been combing through the seeds to find us the finished tasks's cacti. |
Send message Joined: 26 Jun 20 Posts: 25 Credit: 123,735,290 RAC: 66 |
I'm sure this answer is somewhere in the message boards, but, well, i'm old and can't find it. Will there EVER be any more work for Singlechunk population seed cactus stacking? (I need about 40 hours to get my bronze badge at WUProp :) ) |
Send message Joined: 7 Sep 20 Posts: 6 Credit: 2,800,697 RAC: 34 |
The last post was from a gentleman who wanted a specific type of WU's for a valid personal desire. I just want a WU to process. new to the project and have not received my first WU. Bill F In October of 1969 I took an oath to support and defend the Constitution of the United States against all enemies, foreign and domestic; There was no expiration date. |
Send message Joined: 14 Jun 20 Posts: 78 Credit: 1,321,619 RAC: 0 |
Likely, yes. |
Send message Joined: 14 Jun 20 Posts: 78 Credit: 1,321,619 RAC: 0 |
Understood, more tasks coming soon. |