meteokr

joined 1 year ago
[–] [email protected] 2 points 3 months ago (1 children)

I remember the very strange control scheme it had on PS4 I think it was? You couldn't bind your abilities to any button, just specific combinations like Square+Left, but not Square+Right, something like that. I wonder if they've changed that in the newer ones.

[–] [email protected] 4 points 3 months ago

I only just found out about it and wanted to share since I havent heard any dicussion around this style of game. It looks fun, but I havent had a chance to sink much time into it.

[–] [email protected] 4 points 3 months ago

I can't seem to find it, but I think it was James Gosling, where he was blocked from reviewing code at Google because he hadn't gone through the company's approval process. I hope this wasn't a myth I've been carrying on for this long.

 

Looks to be really interesting, similar to the Everquest or WoW private servers with bots, or FFXIV's NPC dungeons, but actually designed from the ground up to be a single player experience where the bots actually play with you. Sounds neat.

[–] [email protected] 2 points 3 months ago

Right, I did hear about that lawsuit way back when, I just didn't know of these types of consequences. Very appreciated, especially the sources.

[–] [email protected] 15 points 3 months ago

I really appreciate you linking studies about this topic, as finding this kind of research can be daunting. Those looks like really interesting reads.

[–] [email protected] 4 points 3 months ago (5 children)

Is this for hardware RAID controllers, or have you experience software RAID like LVM or ZFS exhibiting the same drop out behavior? I personally haven't but it be nice to look out for future drives.

[–] [email protected] 0 points 3 months ago* (last edited 3 months ago)

Drive 1: A, Drive 2: 1/2 A, Drive 3: 2/2 A. Drive 2 + Drive 3 = Drive 1. Hmm that would only be one set of the party though. So you could also add 1/2 of A to Drive 1, and 2/2 to Drive 2 so that the parity on Drive 1 + Drive 2 = Drive 3. Which is extremely silly, and doesn't make a lot of sense to use in the real world.

[–] [email protected] -5 points 3 months ago (2 children)

Oh thanks for the tip! I've edited my comment to reflect the minimum of 4 drives for a RAID6 array.

I've not used RAID6 for a small array like that before so I didn't know it had a conventional lower limit. From the technical sense it doesn't have to have 4 drives, it just wouldn't make any sense to use it that way so I see why software wouldn't support such a use case.

[–] [email protected] 1 points 3 months ago

Yes their failure rates are usually a bit higher, but usually less than the increase in rate from using more than one disk instead. A bit of math can be done using Backblaze's disk failure rate data to get a reasonable approximation of the overall risk of failure.

[–] [email protected] -5 points 3 months ago

Exactly! RAID gives you the breathing room to react to the partial failure of the full RAID array disk. I appreciate your understanding.

[–] [email protected] 6 points 3 months ago

Oh I thought there was some other CVE acronym I was unaware of. I don't think periodically git cloning a repo every few days would be something to worry about. Ever since the Yuzu take down I got in the habit of mirroring a bunch of repos that I'd be very sad to lose, just as a precaution, it probably won't matter, but it's a tiny peace of mind knowing I could at the very least compile it myself if it was lost.

58
submitted 8 months ago* (last edited 8 months ago) by [email protected] to c/[email protected]
 

Seems like a really cool project. Lowering the barrier to entry of locally run models. As llamacpp supports a ton of models, I imagine it be easy to adapt this for other models other than the prebuilt ones.

view more: next ›