Getting an error, just got a new mainboard + CPU, and have downloaded

File: i3_6100_graphics_error2.png (18KB, 618x453px) Image search: [Google]

18KB, 618x453px

>>157474
And second error. Driver won't install. I've already tried uninstalling the old driver, but it won't uninstall.

which CPU and board?

>>157518
Skylake i3 6100
Mainboard h110m-a asus

I'm guessing the on-die graphics chip on the i3 is more powerful than the onboard graphics of the mainboard though, so hoping to get the i3's on-die graphics installed.

>>157604
Looks like that CPU has Intel® HD Graphics 530

>>157618
link:
https://downloadcenter.intel.com/download/26078/Intel-Graphics-Driver-for-Windows-10-and-Windows-7-8-1-15-40-?product=88345

no GPU though?

>>157606
Please go and stay go.

Motherboards have not had on-board graphics in the sense you mean since around 2005.

"Onboard graphics", nowadays, means the motherboard provides a PHY for the CPU to use. It does not mean it has any kind of graphics ability of its own.

Stop trying to be helpy, when you clearly don't know enough to help.

>>157474
OP, your /chip/ is a "6100".

It has built in an "Intel HD 530" GPU, the driver for which you can find by googling "intel driver 88345".

>>157637
fuck off faggot.

>>157474
I see what your problem is, 6100 is the name of your processor model not your GPU model. The Iris 6100 is found on the i5-5257U, which is a laptop chip. As the other anon pointed out, you want the drives for HD graphics 530. Also, if you're running windows 10 then it'll download the drivers automatically in which case your best bet is to just leave it alone. It'll also update the drivers along with the rest of the OS so you don't need to think about that.
>>157637
>Motherboards have not had on-board graphics in the sense you mean since around 2005.
While I agree with your sentiment, there are AM3+ motherboards with radeon 4000 series IGPs. The chipset is older than dirt but the motherboards are specifically AM3+. The 2005 date doesn't hold water either because Intel didn't remove them from the motherboard until Nehalem IIRC, which was 2008.

File: rages_at_legitimate_answers.jpg (105KB, 600x600px) Image search: [Google]

105KB, 600x600px

>>157642
Ooh, salty.

I made this for you.

>>157658
Not him but OP probably did google it with the wrong terms, that's probably what got him in this situation in the first place. He also did give the relevant information, which is why both you and another anon found out what he was doing wrong. I'm not sure I'd call your answer legitimate as you called him an idiot for a pretty common misunderstanding while spewing a ton of incorrect information yourself.

TL;DR, people in glass houses...

>>157668
Explain how
>the driver for which you can find by googling "intel driver 88345"
Is not a legitimate answer?

Every single result for that search is the page on Intel's site with the correct driver he needs.

Sure you're not him? You seem a little booty-blasted.

>>157675
>Doesn't address the fact that he made an idiot of himself
Nice dodge anon, I almost didn't notice it what with you intentionally misreading something.

File: dodge-doge-truck-stickers-wow-1392502862h[1].jpg (662KB, 1563x2047px) Image search: [Google]

662KB, 1563x2047px

>>157676
>nice dodge
Right back at you. You've still not explained how something that completely and perfectly solves the problem can ever not be a legitimate answer.

Answers to dumb questions come with a side of vitriol here.

If that triggers you, there's always a place you can go.

>>157687
>If that triggers you, there's always a place you can go.
Again ignoring the fact that you weren't answering a question, you were replying to an anon who had a misconception and you did it by spewing fucktons of incorrect information. In your own words, stop trying to be helpy.

File: giphy[1].gif (239KB, 300x401px) Image search: [Google]

239KB, 300x401px

>>157698
>not OP
>an anon
You're really having difficulty getting your story straight.

Are you talking about one person (>>157668)? Or two separate people (>>157698)?

Your writing style is distinctive; there's no-one else here who writes like you. There's only four posters in this thread, and two of them are my phone and my laptop.

Install your damn driver, and enjoy your working computer. You're welcome.

>>157687
>vitriol
easy there, fedora neckbeard

>>157704
You're right, my writing style is distinctive. Therefore it should be abundantly clear that I'm not OP, just an anon reacting to you being an idiot and now trying to hide that your an idiot. Funny that you're calling me mad when you're the one shitposting after having been called out.

OP here, finally back from work.
>>157622
Does not work, giving same error as before.

>>157637
>Please go and stay go.
That was my post. I was elaborating on my last post, it wasn't someone replying to me.

As for your comment on onboard graphics, the correct term for what I'm looking for is on-die graphics, because it's built onto the CPU die. Most modern Intel CPU's are in fact APU's.

>OP, your /chip/ is a "6100".
>It has built in an "Intel HD 530" GPU, the driver for which you can find by googling "intel driver 88345".
Just Googled, downloaded, and attempted to install that. I'm certain that's one of the same downloads I've tried yesterday. Same error.

>>157652
>Windows 10
Ew, no thanks. Might get some hate for this, but I legitimately do not want to touch Windows 10 with even a broom handle.

Anyway, am I fucked until I buy a dedicated GPU? This is going to take me a little while to get hold of, was hoping to use the full resolution of my monitor at least until I can buy a dedicated card.

>>158188
>Anyway, am I fucked until I buy a dedicated GPU
You shouldn't be, no. What drivers were installed? What does it show up as in the device manager?

Also, why don't you want windows 10? What version are you using?

>>158188
Nothing is an "APU", that's a stupid marketing term made up by AMD to make their onboard graphics sound more impressive than it actually is.

The "error" you're talking about is Windows warning you that you're installing a driver for a different device than what it thinks the device is.

This is hardly unexpected given you've been dicking about with it.

Intel's compatibility chart for that driver clearly lists your processor and OS, so if you do in fact have the processor you claim and the OS you claim, the problem is between your chair and your keyboard.

>>158202
>that's a stupid marketing term made up by AMD to make their onboard graphics sound more impressive than it actually is.
No it's not and you're an idiot if you think it is. There's quite a lot more to an APU than just having an on die GPU, I suggest you read up on HSA before you spew drivel again.
>>158188
>Most modern Intel CPU's are in fact APU's.
They come close to it but they're not quite there. They don't support all the required features.

>>158211
>There's quite a lot more to an APU
No there isn't.

It's literally just moving the graphics that used to be in the northbridge into the CPU so as different versions can be offered in a customer-replaceable part, thus achieving market segmentation.

There's no technical reason why it has to be in the CPU, and having it in the CPU means every machine with a discrete GPU has a redundant onboard graphics sitting there wasting silicon that could have been more cache or another core.

It's marketing, pure and simple.

>>158231
>It's literally just moving the graphics that used to be in the northbridge into the CPU
No, it isn't, if you'd looked up HSA on wikipedia you would have seen that. There's so much more to it but the main, absolutely massive point is that the GPU and CPU have access to the same memory. That means the GPU can work on the same thing as the CPU without incurring the massive overhead that was associated with the previous way of doing things because previously you'd have to copy it from the memory of the CPU to the memory of the GPU, let the GPU do its thing and then copy it BACK to the CPU. You're an idiot and you shouldn't open your mouth if you don't know what you're talking about.

>>158560
Yeah, no.

Maybe you should google "bus mastering", "concurrency", "memory-mapped I/O" and "cache coherency".

Aside from the fact that the situation you describe has always been the case (shared memory, memory controller in northbridge, "GPU" in northbridge; shared memory, memory controller in CPU package, "GPU" in CPU package)*, there is absolutely no workload where you would ever want two execution units working on the exact same data. Synchronising your caches and waiting on the other execution unit clobbers your performance dead. So absolutely nothing does it. Even if the GPU was just invalidating cache lines all the time, the processor would take a ~90% performance hit, as it's effectively executing from L3. You cannot have processors waiting on other processors; the very first, most basic optimisation is to eliminate this.

Once you have non-retarded code that isn't using two execution units on exactly the same data and killing the system with IPC, then it immediately stops mattering how fast each unit can access data the others have locked. All that matters is that there's sufficient bandwidth between them, and guess what?

There always is. No-one's ever designed a system with an inadequate backplane, because that kills performance dead. WOW FASTER BACKPLANE is the one thing every new architecture has boasted about, from VESA local bus through Hypertransport to QPI.

You just fellated a marketing term with dubious scientific provenance and no real-world relevence.

* Even the original Xbox from 2001 worked this way. As did the Nforce GPU it's based on.

File: HSA_–_using_the_GPU_without_HSA.svg.png (51KB, 720x540px) Image search: [Google]

51KB, 720x540px

>>158565
>Aside from the fact that the situation you describe has always been the case (shared memory, memory controller in northbridge, "GPU" in northbridge; shared memory, memory controller in CPU package, "GPU" in CPU package)*,
Uh huh.
>You just fellated a marketing term with dubious scientific provenance and no real-world relevence.
Just stop with the idiocy and the typos, please. It's a big deal, in fact a large chunk of OpenCL 2.0 is built around it.

Seriously, just go read the wiki article.
https://en.wikipedia.org/wiki/Heterogeneous_System_Architecture

File: image.jpg (76KB, 650x477px) Image search: [Google]

76KB, 650x477px

>>158571
Again, you seem to think this is a new thing.

MMIO and bus mastering is how things have worked, since the original PCI bus.

File: image.jpg (100KB, 800x800px) Image search: [Google]

100KB, 800x800px

>>158577
>>158571
Look familiar?

inb4 herp derp that's different because it just gets on with it and hasn't got a stupid name

>>158577
>>158578
While that's shared memory it doesn't allow both the CPU and GPU to work on the same memory. That's a distinction you seem to be struggling with, and severely at that. Seriously, read the wiki article. If you think you know better than the wiki article, feel free to edit it and correct the books that have been published on the topic and so on and so forth.

>>158577
I looked into the xenos GPU, on the off chance that maybe you weren't an idiot and that maybe you had one good point but no, alas not;
>Procedural Synthesis Technology(XPS): During read streaming into the CPU, a custom prefetch instruction, extended data cache block touch (xDCBT) prefetches data directly to the L1 data cache of the intended core, which skips putting the data in the L2 cache to avoid thrashing the L2 cache. Writes streaming from each core skip the L1 cache, due to its no-write allocation (avoids thrashing of high-bandwidth, transient, write-only data streams on the L1 cache), and goes directly to the L2 cache. The system allows for the GPU to directly read data produced by the CPU without going to main memory. In this specific case of data streaming, called Xbox procedural synthesis (XPS), the CPU is effectively a data decompressor, generating geometry on-the-fly for consumption by the GPU 3D core.
Which is nowhere near the same thing as what HSA enables. You really, really need to read the wiki article.

>>158579
Yes it does.

It's right there in the block diagram. You can see it there yourself.

Seriously, dude, if there was something that let a processor and a GPU share memory at a byte granularity and magically synchronise multiple threads on the same data with no overhead, do you not think someone would have used it to join *processors* together and in one stroke solve every problem related to concurrency? If it actually did what you seem to think it does, then why not just have ten 2GHz processors "work on the same memory" to produce a 20GHz processor?

You're essentially saying "this magic acronym makes the entire field of high performance computing go away and lets us treat everything as one big processor", which is as close as you can get to an outright admission that you don't understand the subject.

File: dragon thing.jpg (177KB, 536x777px) Image search: [Google]

177KB, 536x777px

>>158583
>Seriously, dude, if there was something that let a processor and a GPU share memory at a byte granularity and magically synchronise multiple threads on the same data with no overhead, do you not think someone would have used it to join *processors* together and in one stroke solve every problem related to concurrency? If it actually did what you seem to think it does, then why not just have ten 2GHz processors "work on the same memory" to produce a 20GHz processor?
You're just trolling at this point, right? You're not actually this stupid, right? If so, 10/10, you got me good.

If not, for the umpteenth time, read the damn wiki article and go look up OpenCL 2.0

Also, very nice of you to take a powerpoint for the xbox 360 and an unrelated block diagram from the wiki page for HSA, I like how those two have nothing to do with each other and that you're trying to make them related.

>>158581
It can do all that other stuff *as well*.

What you (bizarrely and irrelevantly) copied is a specific optimisation in a processor for a games console that allows it to run game loops without pointlessly clobbering its cache.

There's absolutely nothing stopping the CPU and GPU communicating through shared memory; this is a specific optimisation that allows them to *bypass shared memory entirely* and pass geometry from the CPU's L1 to the GPU's L2.

You really, really, really need to stop talking about computer architecture.

>>158585
>I like how those two have nothing to do with each other and that you're trying to make them related.
If you can't see the architectural similarity, you're really not qualified to opine on the subject. They very explicitly depict the same thing.

File: Integrated_graphics_with_distinct_memory_allocation.svg.png (37KB, 720x405px) Image search: [Google]

37KB, 720x405px

>>158586
>It can do all that other stuff *as well*.
From the wiki article;
>In partitioned main memory, one part of the system memory is exclusively allocated to the GPU. As a result, zero-copy operation are not possible.
See image.
>>158588
>An IGP is an IGP therefore an unrelated block diagram goes with a powerpoint I found somewhere else
You.. what?

>>158590
Your point?

Neither Xbox nor Xbox 360 had partitioned main memory. PS3 did, which made it a bitch to port to.

Gain a proper understanding of the subject, make your own arguments, and stop copying phrases into Google, finding the first bit of text you find that looks complicated and contradictory, and pasting it in here.

It's somewhat fitting that your last redoubt is copying blocks of data around.

>>158595
>Neither Xbox nor Xbox 360 had partitioned main memory
You're right, the 360 did indeed have a form of UMA. It still does not do anywhere anything resembling the HSA spec. It also does not mean that he GPU and CPU shares memory in the same way as with HSA. Again, read the damn wiki article.

>>158600
No, that's literally what it does mean.

Please just stop.

>>158191
I'm on Windows 8.0

Don't want Windows 10 because Microsoft spyware. Don't want 8.1 because it's likely the spyware they've introduced in Windows 10 is also included in a fully updated 8.1

I went out and bought a CD drive, only to be told by the CD that it's not compatible with my OS. Looks like Asus are full of shit, blocking you from using Windows 8.0

I'll wait until the custom Sapphire RX 480's are out and buy one. Fuck Asus, fuck Intel, and fuck Microsoft, I'll resort to the Radeon solution.

Unless there's a way to trick the CD into thinking I have Windows 8.1 ...

>>158699
Almost forgot to mention, I'm on the resolution 1280 x 1024 and this is a 4K monitor. You can imagine how annoying that is. Funny thing is, when I first plugged the monitor in, I was immediately able to access the full 4K resolution. It was beautiful. Was like that for two days then had to install Java (I fucking hate Java, but some software I use now requires it), and since rebooting, I can no longer access 4K resolution. Java surely wouldn't have cause this, would it?

>>157658
the fisherman casts multiple lines into the swamp. thick waters will make it difficult to pull in his catch, but the swamp is a lively place.

>>158188
didn't read the rest of the thread, but last checks for installing:
looks like you're using windows 10. make sure its the right version, 32/64 bits.
check the BIOS settings, make sure everything looks right
update windows
update BIOS version

if you get a dedicated GPU, and still have problems, then come back here and tell us about it. the default driver apparently works anyways.

>>158699
don't feel like rewriting above.
I'm pretty sure you already fucked up by using windows 8. I recommend using 10 with tinywall. You can use it to block the updater and other windows systems, and whitelist only.
Also, some people said that they had to update windows to make it work with that processor. I'm guessing you don't update windows 8, so that could be a cause.

>>158700
perhaps dependencies or a program that required java did.

>>158701
Thank you for this advise. What you said about Windows 10 sounds like a good idea... what I'm also concerned about though, is Windows 10's ability to track your photos. Don't need my loli pictures to be sent to M$ database, really.

If there's a way to disable this, then I'm all for it. What's the best way to install Windows 10? Do I really have to go the Windows 7 pirate -> upgrade route? I'd rather do a fresh install of an OS natively (pirate Windows 10 directly rather than pirating an older version then upgrading)

Honestly I could just wait for the Radeon cards to come out... but then again, DirectX 12 is a thing, and this thing supposedly only works on Windows 10

>>158703
I'm not aware of any specific microsoft systems, but I don't use any of their built in apps. I removed some of the built in programs for useless features I don't use.
The firewall seems to take care of pretty much everything else. Of course, recklessly disabling the firewall or creating exceptions leads to problems.
Also used CCleaner to manage startup tasks, cut everything I didn't need. Boots in 13 seconds average.

Just install a fresh windows 10.

>>158706
>Just install a fresh windows 10.
How though? Is there a way without resorting to the old 7 -> 10 trick?

>>158692
>No, that's literally what it does mean.
So you think that all UMA is HSA and that HSA is just UMA? Seriously, JUST READ THE WIKI ARTICLE AND THE OpenCL 2.0 specification and you'll see why that isn't the case. Jesus fuck you cannot be this thick.

>>158722
The same way you would install windows 7?
That's how I did it, nothing special. It's an old version, but pretty much the same thing. It's not like updating will prevent viruses as much as this firewall will, unless you like to click on clickbait and download suspicious things.

Most of their updates will probably even more intrusive and unnecessary over time, to get you used to the kind of stuff they forced out in windows 8.

>>158727
I've installed a non-pirated version of Windows 7 once. I went straight from XP to 8