ERMAGERD ARGH GODDAMNIT! FERK YU IN DA BUMWHOLE!!
Er… I mean… hi. How are you? 🙂 Don’t worry, that was just me raging at DAZ Studio as per usual. 😛 But because the reason for that has been mentioned about 6000 times before, let’s start somewhere more interesting first. 🙂
Pokémon Super Mystery Dungeon
Well, Amazon failed massively and said they weren’t going to have it in stock until after the weekend I’d been planning to play it. I bitchslapped them and cancelled the order (didn’t really bitchslap them but I did IN MY HEAD), and ordered it from ShopTo instead. It cost a tiny bit more but it arrived on release day, so GG to them. 🙂
How is it? Well, I’ll do my best not to throw any major spoilers. 😛 At first, I wasn’t really “feeling” it. The writing and the dialogue was great as usual, but the setting felt a bit wrong. You and your partner are actually children this time around, which felt strange for some reason. The first time my partner appeared, I thought “Oh God, please don’t tell me Chunsoft have made him an irritating little arse this time”, but after playing for a while, I feel like they introduced him in that way for a reason, so that’s okay. 🙂 There are a couple of other things that are still bugging me, though.
First off, the story doesn’t actually go anywhere for ages. Sure, a story is there, but it’s just a series of unrelated events and doesn’t really mean anything. Something happens, the thing is resolved, normality resumes for a while. Every so often, there were hints at this giant, overarching plot, but you never seem to get involved in it. It feels like the writers got stuck in first gear. When you get past a certain point, though, it really starts to open up. When you battle a Legendary Pokémon for the second time, shit gets REAL, with sparkliness and henshin all over the place. 😀 That was one of the few times that battles in Mystery Dungeon games were genuinely exciting.
The excitement probably came from the difficulty… the noticeably increased difficulty. In that battle, the boss loved to use ranged attacks which did a lot of damage, and I was forced to think between almost every turn, adjust tactics, consider item usage and decide on a move. It felt tense, like a true roguelike. 😀 Overall, it feels significantly harder than all its predecessors. That’s both a good thing… but kind of a bad thing as well, because I’ve found that the difficulty curve is totally erratic. Very early on in the game, I accepted a job to beat a Salamence, which had a rating of one star, the lowest possible. When entering the floor where Salamence was, he wiped my team so quickly that I thought I’d made a mistake when picking a mission or something. Oh, no, I hadn’t. His stats were just friggin’ stupid. I could only do single-digit damage to him, even with my strongest moves, and he could one-shot me. I had to resort to dirty evil tricks to beat him. I brought along a wand item which causes an enemy to fall asleep, and kept using it on him every time he woke up, spamming Alliance Attacks while he was asleep. Even that wasn’t enough, so I used a Blast Seed to compensate for my lack of attack power, which is cheap, since it always does 100 damage (quite a lot, especially in the early game). After beating him, I actually had the option of using him on future missions. I checked his level and found out he was Level 50. I wasn’t even Level 20 yet. 😛
You see… instead of recruiting Pokémon by beating them in a dungeon, this time around you “recruit” them when you complete a job that involves them. It treats each Pokémon as an entirely different character, and that’s nice, but the problem with this that they’re not necessarily close to your level, and that massive lack of balance is my biggest gripe with the game so far. If the job involves beating them, and they’re say, Level 50, it’s impossible without using cheap tactics and overpowered items. And then, when you DO win, you have an overpowered team member you can use at any time! And it’s not like all the jobs require you to beat an overpowered enemy, either; one job was just “take me to this Pokémon”, and after completing it, I got the ability to put a high-level Dragonite on my team! I mean, if I really wanted, I could easily make a team consisting of a Level 56 Alakazam, a Level 55 Dragonite and a Level 50 Salamence and steamroll the shit out of everything. But that’d be like playing with God Mode on, and that would make the game pointless! Why does it even let me do that? It’s not fun!
Other than it having basically the worst balance I’ve ever seen in gaming, Super Mystery Dungeon is pretty good so far. Its strong points are the dialogue, the story and the characters as usual, but the actual gameplay is probably better than its ever been thanks to the return of hunger (which now does triple damage, so you can’t even try to ignore it any more) and the introduction of the Looplet/Emera system. Equipment items have all changed to Looplets, which have similar effects to the old equipment items (they can increase Attack, Sp. Attack, Defence, etc.), but Looplets have sockets into which you can place Emeras (basically gems), which have various effects. You can also use Emeras you don’t want/can’t use on party members to increase their stats until the end of the adventure. The Emeras disappear when you leave the dungeon, so you only have the Emeras you find in the dungeon at your disposal. It just makes equipment way more interesting. The dungeon crawling can still be repetitive, so I understand why some may not enjoy it as much, but because it’s more difficult than Gates to Infinity, it revives that sense of tension from Time/Darkness/Sky. It’s difficult to compare it to the other games overall since I feel like I’ve barely scratched the surface of the story yet, but because it had a slow, almost uninteresting start, I doubt it’ll be better than Explorers of Sky. The writing in that game was unbelievably good. 😀
OH, OH OH, a couple things I forgot to mention. One: the most disappointing, saddest moment of this game so far:
NOOOOOOOO!! I wanted to spin the V-Wheeeeeeeel!! 😦 Yes, the V-Wave system isn’t in Super Mystery Dungeon, which is a shame, since I quite liked it; it encouraged you to take different team members along all the time. This game tries to encourage that in a… less good way. Basically, at random points, it picks three random team members who become “motivated”, and they gain triple XP from enemies. No Victini or any spinning of wheels or anything. Yawn.
Anyway, second… you know the boss battle theme? Maybe not. 😛 Anyway, it’s been the same melody in every Pokémon Mystery Dungeon game. It sounds like this in Super; it’s quite a “playful” version, which is apt. 😀 Now… unless you’ve reached Chapter 12, the next link is a spoiler… but at a certain point in the game, it gives you the the boss theme as you’d expect, but it’s an alternate version that came pretty much out of nowhere. And it nearly made me explode. I had to reload my save data and redo the boss battle just so I could hear that track again. 😛
Okay, next topic! Well, as we’re on the subject of Pokémon…
Pokémon Omega Ruby
“Wait wait wait wait wait what the fuck who now WHAT?!”
No, I don’t get it either. 😛
I don’t particularly know what it was. Maybe it was my hype from Super Mystery Dungeon. I tend not to get hyped about games any more, so it’s quite special when it happens. Or, maybe it was because of the demo I played. Or maybe it was the constant exposure to Pokémon-related games on the 3DS, because basically every not-actually-free-to-play game Nintendo makes seems to be Pokemon themed. Pokémon Shuffle, Pokémon Picross, Pokémon Rumble World… it’s like Nintendo are using one of the most famous game franchises of all time as a guinea pig for their annoying interest in not-actually-free-to-play mobile gaming, simultaneously promoting the series and ruining it. 😛
Whatever it was, a few weeks before Super Mystery Dungeon came out, I decided to buy Pokémon Omega Ruby.
So… why don’t I get it? Well, my only real interest in the Pokémon franchise has only ever been Mystery Dungeon; a spin-off. I’ve never liked the core Pokémon games. Ever. I had either Red or Blue on a very illegal multicart way back when I had a Game Boy (way back when I was young enough not to know it was illegal :P). Almost everyone else I knew was into Pokémon; the games, and the anime, and the cards… but I got really bored of the game, I hated the shit out of the anime and didn’t really understand the appeal of the cards. I guess that was early evidence of my “fuck the flock; I’m going my own way” nature. 😀
Someone I know had Colosseum on GameCube, and I remember borrowing it for a while. I remember liking the Double Battles but thought the game was even more boring than usual, and it had a stupid difficulty curve, and it had awful production quality. Some number of years later, I played SoulSilver on an emulator, and referred to it in this very blog as “Poké-Way-Too-Tedious-Brain-Numbingly-Boring-And-Repetitive Soul-I-Don’t-Know-WTF-To-Do-Or-Where-To-Go-Next-Silver”, which… basically sums up my experience with the game. 😛 I also played White, and I SWEAR I wrote about it to some degree in this blog but I can’t friggin’ find it at all. I’m pretty sure I said it played exactly the same only it was in Almost-3D and the Pokémon were animated. And by “animated” I mean “they twitched slightly”. 😛
The annoying thing is, I like the Pokémon. Not as much as Digimon, because Digimon tend to be 120% fucking cool, hardcore, unrestrained badasses, like turtles with massive hammers or werewolves that kick your face off or dinosaurs with
rocket launchers GIGA CANNONS which turn into bipedal dinosaurs that throw suns at you, but I still like a lot of Pokémon. 😛 I can remember a fair few species and their types, mainly from playing Mystery Dungeon, and I mostly know what types beat what, so by memorising this stuff, I’m obviously interested. If I didn’t give a shite, I wouldn’t have a clue that Flying beats Fighting, Grass beats Water and Dragon beats itself. 😛
So… well, seeing as how I have a 3DS, and because (real) Pokémon games are exclusive to Nintendo systems, I decided to download the *deep breath* Omega Ruby slash Alpha Sapphire Special Demo Version *whew* and give it a chance, even though core Pokémon games have failed to interest me on four previous occasions. 😛 I don’t really understand why, since the demo was basically devoid of content, but there was something about it that made me think “Hmmm, yeah, this is good… I kinda like this”. So I decided on Omega Ruby, and bought a copy. It’s funny… Pokémon’s basically the only game that you need to research which version to buy beforehand, since there are 22 Pokémon that become incredibly difficult to get if you get one version over the other. 😛
Soooooo? Did I do something stupid? No, actually. Somehow, against all odds, it’s made me a fan.
I don’t really understand why, because it’s basically the same shizzle as the previous games. It’s a cockfighting simulation where you play as an evil kid that forces monsters to beat up other monsters to the point of fainting before you capture them in a tiny ball. 😛 Okay, no, without trying to apply the laws of the real world to the Pokémon world (because then the Pokémon world would seem royally messed up), you’re a Pokémon trainer that goes on a journey with Pokémon, catching new ones and raising them (and being nice to them! really!) to challenge Gym Leaders all over the region with the hope of getting all the Gym badges, then beating the piss out of everyone in the Pokémon League and becoming the ultimate winner of everything.
The battle system is the same. Choose a Pokémon, beat your opponent in one move, receive a microscopic amount of experience. 😛 Okay, maybe not, although to be honest that is pretty much what I remember from the Game Boy games, but I imagine I was doing it wrong and battling Pokémon that were too weak. The battle system is fine, although it’s fairly luck-based and memory assists you more than skill. If you know what type beats your opponent, switch to a Pokémon with moves of that type and you’ll win. If you don’t have that kind of Pokémon, switch to one that resists the moves you think your opponent has but ensure you can still do standard damage with your own moves, and you’ll win. 😛 The two biggest differences are the addition of a new type, Fairy, and Mega Evolution… although both of those were actually introduced in Pokémon X & Y. 😛 Mega Evolution is like a dramatic-looking power boost you can trigger for specific Pokémon, but you can only do it if the Pokémon is holding the required Mega Stone, and you can only use Mega Evolution once per battle.
Even though the battle system is mostly the same, the graphics are a lot… lot prettier. The Pokémon are 3D instead of using 2D sprites, with a cel-shaded anime-like art style. They look great; it’s probably the prettiest game I’ve seen on the 3DS. 😀 The animation’s great, too, and so are most of the special effects. All of this makes battles way more interesting to watch. I never got tired of seeing my Torchic use Scratch; watching him run back after attacking was funny and pretty cute. 😀 Unfortunately, the graphics might be too great. The game limits stereoscopic 3D to Single Battles, some cutscenes and a few other areas… and the reason it does this is pretty obvious. The 3DS just can’t handle it. 😛 The framerate in Single Battles with 3D on is just… wow. It chugs like hell at certain camera angles, usually when both Pokémon are visible or during certain attacks. When only one Pokémon is being rendered, it tends to be fine. I don’t know what the poly count of the models are, but I reckon they’re a bit too high for the 3DS, or maybe the post-processing is too expensive. This is even worse when you keep in mind that I have a New 3DS, which has a bit more power than the standard one! You can massively boost the FPS to about 60 simply by turning off stereoscopy, but… then… then everything looks flat. 😛
Exploration is mostly the same; if you walk in tall grass which covers a large amount of the land, you get attacked by monsters at a rate that makes you wonder how anyone gets anywhere without dying. 😛 It’s slightly better now, though… the encounter rate isn’t quite as bad as I remember, plus in most areas you have full analogue movement, so you can sometimes avoid patches of grass when you don’t want to battle by going diagonally (WOW DIAGONALS). Of course, you can still use Repels to stop ’em from jumping you, and you can also sneak to reduce the encounter rate, apparently. Another difference is a feature called DexNav, which puts sneaking to further use. It’s useful normally because it tells you if there are any Pokémon you haven’t caught in the current area. But sometimes the game will randomly pick patches of tall grass and place a Pokémon there, rummaging in the grass, usually with its arse sticking in the air. 😛 When this happens, you can use the DexNav to scan it, which, when you get close enough, gives you a silhouette of the Pokémon and some info about it. Sometimes it generates interesting Pokémon with unusual moves or abilities, or unusually high levels. The more Pokémon of that type you fight, the better the info becomes, eventually even telling you a summary of the generated Pokémon’s potential. If you want to battle that Pokémon, you just sneak over to the patch of tall grass it’s hiding in and make contact with it.
Catching Pokémon is business as usual, too. You hit it until its HP gets low to increase the catch chance. You can also cause status conditions to increase the chance further. When you’re feeling lucky punk, you chuck Poké Balls at it until the Random Number God smiles upon you. Sometimes this is easy and happens in one attempt, other times it’s annoying as fuck. Like recently, when trying to catch some Pokémon quite some distance from the home town. Sheesh, Skarmory was a bugger. 😛 After about 5 attempts I thought “this is weird”, and looked up the catch chance, finding out that the base chance (with the Pokémon at full health and using a standard Poke Ball) is… 3.3%. I decided not to give up, though. Lobbing basically every ball I had at it while trying to keep my team alive was somehow both frustrating and exciting at the same time. I caught it eventually, just when I was getting low on balls. Heheheheh shut up. 😛
Since this is technically a remake, the story is apparently very similar to Ruby/Sapphire’s story. Makes no difference to me because I never played Ruby/Sapphire. 😛 Anyway, the story is incredibly stupid. There are two Super Evil Teams, Team Magma and Team Aqua. Both came out of nowhere, both have seemingly infinite members, both utterly suck at Pokémon battles, and both want to do something retarded that makes no sense and is impossible. Magma wants to expand the land, and Aqua wants to expand the ocean. So both want to do something they think will be of benefit, but they’re blind to the fact that by succeeding, they’ll probably mess up the planet and kill everyone. 😛
Some of the dialogue is funny, but the script is annoying as fuck, as usual. Everything’s decided with Pokémon battles. So if you want to get something back from Team Magma or Team Aqua, who are basically terrorists and extremists, you challenge them to a Pokémon battle. If you win, they’re all like “OHMIGAWD YOU’RE SO STRONG FINE HERE HAVE THIS THING WE REALLY NEEDED TO DO OUR EVIL DEEDS” instead of “lol you won, so what, Falcon Kick, Mongz blacked out”. Especially, it’s really frustrating when the Evil Team Of Terrorists have just acquired The Plot Coupon To End All Plot Coupons right in front of you, and you just let them laugh at you and walk past. ARGHH! STOP THEM WITH A DICK PUNCH FFS!!
And where the fuck are the police?!
Aside from all that, though, it’s actually an interesting game to play. The world is pretty big, and there are quite a lot of things to do other than battles. There are Contest Spectaculars, which kinda work like a battle, but the tactics are different. You’ve basically gotta get as many hearts as possible in five turns while trying to reduce your opponents’ hearts. There’s Pokémon-Amie, too, which lets you play with your Pokémon to a certain extent; you can feed them, pet them and play a few minigames with them. The minigames are weird and feel like they were made in about two days, and the facial recognition is completely dire. The crap facial recognition ruins a somewhat interesting game you can play where your Pokémon asks you to make a certain face. It’s cool until it asks you to make a huge smile or open your mouth wide, because the camera cannot see me doing it. Then it times out and the Pokémon gets all sad and I’m like “hey sorry, but the facial recognition is total bullshit!” and that’s bad because then I’m teaching my Pokémon bad words. 😛 You can also do this thing called Super Training, which lets you increase your Pokémon’s EVs, or “Effort Values”, officially known as “base stats”. The Super Training minigame is really weird and I don’t get it at all; you throw balls at a puppet in a 3D environment and try to get the 3D balls into 2D nets while avoiding balls thrown at you. If you get enough balls into nets… you, er… you win. And your Pokémon’s base stats increase. There’s also a different kind of training where you give them a training bag and have them break it to get various effects, usually more base stats. They hit it once a minute automatically, but waiting for them to break the bag themselves is utterly stupid and pointless because you can get the effects of the bag much, much faster by tapping the bag yourself. It’s nice that they’ve exposed EVs like this and it’s great that you can increase them more easily, but it’s just a really weirdly themed, badly designed minigame that makes me think they ran out of ideas.
But yeah, overall, I’m enjoying it! Wow. Feels weird, suddenly finding out I like something I didn’t like before. 😛 Hell, I even bothered to go into Brighton (and walking around Brighton is one of my top ten most hated things) and go into an actual GAME store for the first time in years to ask for a code for a Mew during the distribution event they were having in February. That felt like it was gonna be the most embarrassing moment of my life until the cashier asked me how many I’d like, to which I responded “er, just the one, I think” in a confused manner. He then mentioned that the current record was ten. That made me feel much better. 😛 I mean, ten? Why? I thought you could only redeem one code of a kind per save file. Unless you’ve got X, Y, Omega Ruby and Alpha Sapphire and have 6 friends who can’t be bothered to go to GAME? 😛 Or maybe they were giving them out to people in other countries who couldn’t participate? Maybe that’s it.
I’ll probably end up getting either Pokémon Sun or Moon when they come out, and if Omega Ruby dries up in the meantime I might even get Pokémon X/Y. Heh… this is mental. I’m actually spending more time playing on the 3DS than on the PC recently. 😛 I now have 3 RPGs to finish, and I bought Mega Man Legacy Collection too. 😀
What’s next…? Ah yes, here comes the thing I was going to talk about originally; a BIG wall of text about a problem that was stupidly hard to debug! 😛
Digital ASS Zone POOdio and the Chronicles of the Corruption of the Extended Memory POOl
I’ve had these really annoying problems with DAZ Studio for some time now. Problems where during a render, DAZ Studio would crash the graphics driver (the effect of which crashes DAZ Studio if a viewport is visible), or problems where I would get a DRIVER_CORRUPTED_EXPOOL bluescreen without warning. Recently, it got to a point where I was absolutely friggin’ sick of it. It was making most renders impossible. I had to know what it was, and I had to fix it, otherwise I would explode in a fit of rage, and doctors tend to say that exploding isn’t good for your health. 😛
It literally took weeks to figure it out. At first, I poked around and did all sorts of shit in an attempt to debug the problem, and quickly came to at least one conclusion. And it was a very funny conclusion.
I was certain that this error wasn’t caused by DAZ Studio. Heheheh, oops! 😛
How did I come to this conclusion? Well… I did a stress test on my CPU for the first time in ages using IntelBurnTest. It’s one of those programs which are designed to hammer the balls out of your CPU by asking it to compute something stupidly complicated. If you can run the test for a prolonged period of time without your PC crashing or the CPU making a mistake, you can consider your CPU and its current overclock stable. I ran IntelBurnTest on my then-current overclock, and it crashed the system in basically the same way as DAZ Studio does. I disabled my overclock and ran it again. It succeeded. Easy! Obviously my overclock was unstable. Case closed. The next day, DAZ Studio crashed the PC again, so I ran IntelBurnTest again, and it, er… it produced an inconsistent result. That basically rang every alarm bell. 😛
I disabled Turbo Boost, an Intel feature which auto-overclocks your processor when it’s being strained. Test failed. I disabled EIST, aka: Intel SpeedStep, another processor feature which dynamically clocks your processor depending on its current load, mainly to save energy. Test failed. Disabled XMP, a feature that lets my memory run at the speed it was designed to run using parameters specified by the memory manufacturer. Test failed. Then, I uninstalled my graphics drivers, both dedicated and integrated, and tried again. Test failed!
It was failing at stock speeds with absolutely no overclocking or underclocking at all, manual or automatic, and even though although Windows was blaming the drivers (and according to reports I got from analysing memory dumps with WinDbg, various programs such as Logitech Gaming Software, TeamViewer and System (lel)), it appeared that they were only messing up because the CPU made a mistake. I didn’t want to believe it, but it was looking like a hardware issue. I was hoping it was just bad cooling, but my temps rarely went above 80C at full load, which, in my previous experience, is totally fine.
But then… I ran the test in Safe Mode. It succeeded. Believing it was a fluke, I ran it in Safe Mode again for an extremely long period of time while I went off and played some Super Mystery Dungeon. I checked on the PC before going to bed and it had passed about 150 times in a row.
I decided to stop it, exit Safe Mode and run the same test overnight. Found out this morning that it had failed on the third test, less than five minutes after I left it to run overnight. I left it to run again in the morning. I checked it while I was at work, and it had failed on the 33rd test.
After all that, I didn’t think it was a hardware issue. It must have been a software issue, either a driver or some other program that runs at startup was somehow messing up the CPU, causing it to mess up the GPU drivers and cause a crash. But what could possibly be messing up the CPU?!
Then I realised there was a vague pattern. In Safe Mode, you don’t have networking unless you ask for it. So, I decided to unplug my Internet connection. I ran the test for a while. It passed. I installed my display drivers again. It passed the test. I then plugged in my network cable, restarted the PC, and started the test again. It failed on the 11th test. I unplugged my network cable, restarted and did a massive test overnight. Checked it this morning. It had friggin’ passed all 200 tests.
That basically told me that the problem could have only been one of two things. It was either my network drivers, or my firewall. I was hoping that it was my firewall, so I disabled the firewall by not letting it run at startup, and did a test. I checked after an hour or so at work… the test wasn’t running, so the PC had clearly crashed. I tried again, and checked at lunch. It hasn’t crashed, but it had failed on the 6th test. This… is not good, because that indicates the problem lies with the network drivers, and I need these network drivers. No driver, no Internets; simple as.
The only obvious thing that came to mind was the possibility that the drivers were faulty or out of date. I mean, I was having strange problems with them during the Windows 10 Tech Preview… and by “strange problems” I mean “they didn’t work at all past a certain build”. 😛 Of course, my deduction could have been way off too, but… no, it couldn’t have been that far off. It had to be something to do with a thing that accesses the Internet, because when I cut my connection to the Internet, the problem didn’t occur.
I went on Killer’s site and got their latest drivers, which had actually been updated just a couple of weeks ago. Through some miracle I actually managed to install them through TeamViewer without losing connection. I then restarted the PC, and ran another IntelBurnTest, firewall still not running. It passed all of the 100 tests I set. I re-enabled my firewall on startup, restarted, and ran the test again, basically expecting it to pass all the tests so I could finally put this shit to rest. It CRASHED THE PC!
“What the fuck” is all I could think of. I was beginning to believe that I was way off, or that this problem was pretty much beyond my comprehension. Maybe the network drivers were fundamentally broken on Windows 10? Maybe the problem was partially caused by the network drivers and partially caused by the firewall? Maybe I’d been led to believe that it was a network issue when it wasn’t? ARGGGH!
I did the test again thinking it was a freak accident that would never happen again, and TeamViewer lost connection after about 5 minutes: crash! Despite previous tests showing that the firewall had made no difference to the stability whether it was on or off, I disabled the firewall, restarted and ran the test again. Of course it still crashed. I decided to uninstall the firewall ENTIRELY, restarted, and tried the test one more time. Yes, it crashed. And oh-so amazingly and conveniently, the PC didn’t restart after that crash.
So, it couldn’t have been the firewall. But just… all of a sudden, for some impossible reason, it had entered this pattern where the CPU wasn’t failing the burn test by producing incorrect results… it was failing by crashing the PC. The ONLY thing it could possibly have been was the network drivers, but why the fuck would some network drivers screw up the CPU to such an extent that would cause it to make a calculation error?!
Because that crash took out the PC seemingly entirely, I didn’t have a way to confirm whether the network drivers really were causing the problem until I got home. When I did get home, it turned out to be an entirely new type of crash; one that locked up everything. I tried the test again and the same thing happened. So the crash had changed styles again… great.
I uninstalled the network drivers. Ran the test again. And against everything I was expecting… it crashed. This basically ruined the days of debugging I’d done. It was as if the times it passed the test were just unbelievable flukes. Even running it in Safe Mode with Networking with the network drivers uninstalled was crashing. Uninstalling my graphics drivers again didn’t help; it didn’t crash the PC, but it still failed the test.
I noticed something else, though. When running an IntelBurnTest using a small amount of memory, say, 1GB, it didn’t fail. That suddenly made me think it was a memory issue, so I decided to run Windows’ memory checker tool thing. Amazingly, it actually found a memory issue. It was totally unhelpful and just said “oh there’s a problem contact the PC manufacturer”. Uh… I built the thing, so technically I’m the PC manufacturer. 😛
But that test finally made me think “shiiit, that’s gotta be it!” I started to shuffle my DIMMs around to test whether there was a problem with one or both of them, or if it was a motherboard problem. After finding out that the memory test failed with both DIMMs in regardless of the slots they were in, I tried putting one DIMM into a specific slot. The memory test failed, so I replaced it with the other DIMM. The memory test was successful. I left that DIMM in and did a huge burn test using the same memory setting as usual.
And it failed. Crashed the PC after about half an hour.
So… to sum up so far. My PC was not overclocked in any way and my CPU never exceeded 85C even at full load. Despite this, and only when the CPU ran at full load, the PC would crash in one of three unique ways when my network cable was plugged in, but when it wasn’t, it never crashed, but uninstalling my network drivers or firewall made no difference, and it didn’t matter whether I was in Safe Mode w/ Networking or not, but it never crashed in basic Safe Mode. There was apparently a memory issue, but when I resolved the memory issue by removing the allegedly faulty DIMM, it made no difference to whether the crash happened or not.
What the fuck. WHAT. THE. FUCK.
I totally lost my patience after this and didn’t bother trying to debug the problem for a few days. It became an annoying weight on my mind, so when I came back to it, I tried a couple of other things. I updated my BIOS and every driver that came with the motherboard. No effect. I re-experimented with my DIMMs and confirmed that one of them was faulty, and removed the one I was absolutely sure was faulty. Still no effect.
I collected my thoughts on the state of my hardware. I assumed it wasn’t a CPU issue, because if it was, the PC would crash all the time, not just at full load. It wasn’t a heat issue because the same test passed all the time in Safe Mode when the CPU was running at the same heat. It wasn’t a memory issue anymore because I’d removed the faulty DIMM, and memory tests were passing all the time with the current working DIMM. It wasn’t a power issue because the stated max power drain of all my components didn’t even come close to 850W. It wasn’t a GPU issue because the GPUs weren’t in use during the CPU test and the crash occurred when the drivers for my GPUs weren’t even installed. It wasn’t a drive issue because this bug had nothing to do with the SSD or hard drive. And although I didn’t have enough info to clear my motherboard of suspicion, I didn’t believe it was that.
Thinking it through, I was 100% convinced it was a software issue, but it definitely wasn’t the GPU drivers. Because I had no clue besides that, I went into yob mode and decided to disable or uninstall all suspect drivers and programs. I’d uninstall a thing, restart the PC and then do the test. So… I uninstalled Virtual Audio Cable, a driver that always felt really intrusive to me. It had no effect on the test result. I went through my startup programs list, disabling them one at a time. No effect, even when they were all disabled.
Then I realised PlayClaw wasn’t in the startup list. That’s because I have it as a task in Task Scheduler instead of a startup program, because it’s really annoying and asks you for admin rights when the PC starts if you have it as a traditional startup program; by setting it in Task Scheduler you can give it admin permission there so it doesn’t bug you. I disabled this task, restarted, and did the test.
The test passed. It was obviously a fluke, so I did it again. And it passed. And again, this time for 10x longer. Passed.
“Fuck off, stop screwing with me” is what I said. I decided to give it the test it would fail without a doubt; rendering my newest scene in DAZ Studio. In that scene, I use the preset lighting supplied with the environment I’m using. It’s pretty, but like the majority of preset things in DAZ Studio, it’s so expensive and idiotic that it takes my PC around 5 hours to render it out at max quality, and because it takes that long, it has never completed successfully. I set it to render overnight.
It succeeded. IT SUCCEEDED.
I’m not gonna show you the render because there’s so much wrong with it (mainly lighting issues, heh, why the hell did I bother with that preset crap), but what the hell… how?! How could PlayClaw, a game recorder, cause this problem? It’s never caused any problems like this when I actually use it! Christ, it’s not even in use when the problem happens! It’s just that the CPU is maxed!
At first, I thought that the problem might not have been PlayClaw alone. I thought it was a combination of PlayClaw and that faulty DIMM I removed, but… well, for fun, I put my faulty DIMM back in and it still hasn’t crashed, even after several successful burn tests and another render. So maybe it was just PlayClaw. But… argh, HOW CAN A GAME RECORDER MESS UP THE CPU?!
I’ll have to see how it goes. But DAZ Studio is working perfectly now! Er… well, no. It was working perfectly the entire time. Well, okay, not perfectly, because DAZ Studio still has a bunch of bugs, but at least it can do what it’s designed to do now. 😛
As for that render, I’ll have it soon enough. As I mentioned, the render time is utterly insane, even on draft quality. But I have a pretty good feeling about it. 😀
RIGHT, that’s enough. That’s definitely enough. 😛 I’ll leave my PC to render this out and hope it doesn’t crash while I faceplant on my bed. NIIIIGHT