This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
Messages - willrandship
16
« on: August 12, 2014, 09:57:35 pm »
In my experience by the time you upgrade you want to replace the whole thing anyway, but then again, I am a cheapskate. I say go for the H81M.
17
« on: August 11, 2014, 06:19:14 pm »
That's really not simple at all for any terminal-based language. Input for those languages is not handled in real time. Imagine making a basic snake with only input, not getkey, and you'll be pretty much spot on.
18
« on: August 11, 2014, 05:57:44 pm »
What are the rules on how input must work? All the languages I've been using so far don't support raw input without libraries, and it would be faster to do it via line based input, but also much worse.
It will also be impossible for you to test without installing perl/python and whatever library I use.
19
« on: August 07, 2014, 06:39:31 pm »
Besides, it's easier to install smaller motherboards in bigger cases. The screw holes will still line up, and you'll have more room to work with.
20
« on: August 07, 2014, 01:36:21 am »
Code golf shouldn't interfere with the cage matches. It's open to wide entries, weekly, and doesn't take much time, whereas cage matches will be locked between two people for any given match, and have a longer timespan. I think it would be easy for them to coexist.
21
« on: August 06, 2014, 03:33:40 pm »
While that is true, it is less useful than it sounds for single-core performance. All modern OSes I know of use RAM pages, which allocate* blocks of the sticks of RAM for usage in program. If you need an address outside the currently allocated space, it performs what's called a page fault (not an error, more like a garbage collect) and allocates the needed block. This means the program has to wait until that block is allocated.
The only case where dual-channel memory helps in this scenario are TWO programs getting a page fault at the same time. Not just two programs using memory, but two programs happening to need non-allocated memory at the same time.
On the other hand, as far as page fault time usage is concerned, Linux uses about 30% of its CPU time dealing with page faults. So, the performance difference for multiple RAM-heavy threads will be noticeable, but it won't be double. It might be 8-10% faster in that specific scenario.
One stick of 8GB RAM is cheaper. Is the extra cost worth the slight bit of extra performance, and the inability to do a partial upgrade later? (ie buy another 8GB stick and get 16GB total, rather than buying 2 8GB sticks later to swap out the 2 4GB sticks)
*in this case, allocation refers to either copying the block of RAM into the CPU cache, or letting the MMU do some magic. It's not referring to malloc() or new() stuff, although those can easily cause page faults.
22
« on: August 05, 2014, 07:58:22 pm »
75 bytes of perl submitted
23
« on: August 05, 2014, 07:52:33 pm »
If they're separate files, you should be able to send them individually, then archive them in turn. ie Send, Archive, Send, Archive. I don't think the CLI interface has send-to-archive as a feature, unfortunately.
24
« on: August 02, 2014, 06:08:36 pm »
I don't know about pressing enter after every key. I'm pretty sure I specified a string for input...  Better pull out that sub( command. Aligning to the right is fine, as long as it's vertical. FWIW I have a 75 byte perl version.
25
« on: August 02, 2014, 01:22:04 pm »
Well, I'm going to be gone for a few days, so I'll post my suggestion for the next one now. I'm assuming no one else is going to post a perl solution.
It would go as follows:
For a given string input consisting of only uppercase letters and numbers, add the ASCII value of each alphabetical character (that's 65-90 for uppercase A-Z) and subtract every number. Display the result, but printed vertically with each digit on a new line.
You CAN end up with a negative number, in which case the first line should have a - sign.
Example:
Input: "A45FTUX"
Internal Math: (65 - 4 - 5 + 70 + 84 + 85 + 88 = 383)
Output:
3
8
3
A negative example:
Input: 99A874512995
Internal math:(-9-9+65-8-7-4-5-1-2-9-9-5)
Output:
-
3
26
« on: July 31, 2014, 02:39:30 am »
I love the idea. It would really be fun to do a printed version, too, although it defeats the purpose of a calc game a bit.
AoC recommended some great people and projects for the colors. I'd try to get myself a spot, but I don't really think I deserve one.
27
« on: July 31, 2014, 02:30:22 am »
Yes, assuming >=2 is totally fine. IMO, code golf should be about handling the meat of the question, not about handling all the wacky edge cases. If ignoring edge cases makes a smaller solution, that's what I want to see.
28
« on: July 28, 2014, 09:10:22 pm »
Now everything's more complicated :\ stupid 2-byte tokens.
29
« on: July 28, 2014, 07:34:47 pm »
By the way, if your language does not support underscores (ONLY if it does not) you may use a decimal [.] symbol instead.
30
« on: July 28, 2014, 01:19:04 am »
Here's my suggestion for the next challenge:
Input: a positive integer of reasonable size
Output: The highest prime factor of that integer, in binary, but with all 0s replaced with underscores (_) and all 1s replaced with minus signs (-)
Example:
Input: 15
Output: -_-
Example:
Input: 7
Output: ---
Example:
Input: 115
Output: -_---
|
Show Posts