This sounds pretty plausible. The windows user is the least likely to understand the implications of arm for their applications in the ecosystem that is the least likely to accommodate any change. Microsoft likes to hedge their bets but generally does not have a reason to prefer arm over x86, their revenue opportunity is the same either way. Application vendors not particularly motivated yet because there’s low market share and no reason to expect windows on x86 to go anywhere.
Just like last time around, windows and x86 are inextricably tied together. Windows is built on decades of backwards compatibility in a closed source world and ARM is anathema to x86 windows application compatibility.
Apple forced processor architecture changes because they wanted them, but Microsoft doesn’t have the motive.
This has next to nothing to do with the technical qualities of the processor, but it’s just such a crappy ecosystem to try to break into on its own terms.
Decent only if you look at raw performance for the price compared to other MSRPs.
When you scratch beneath the surface a little and see what they’re having to do to keep up with the 3 year old low end Nvidia and AMD parts (that are due to be replaced very soon), it paints a less rosy picture. They’re on a newer, more expensive node, use a fair bit more power, and have a larger die size by quite a bit than their AMD/Nvidia counterparts.
Add to that Intel doesn’t get the discounts from TSMC that Nvidia and AMD get, and I’m doubtful Battlemage is profitable for Intel (this potentially explains why availability has been so poor - they don’t want to sell too many).
While it’s true the average buyer won’t care about the bulk of that, it does mean Intel is limited in what they can do when Nvidia and AMD release their next generation of stuff within the next few months.
If I can get one I’m buying one. I think their performance/cost ratio is excellent, and will probably make NVidia and AMD bring down their mid-range card prices.
But I’m not forgetting who made the prices come down. I’m all in on supporting a new player in the GPU game, and the 5060 would have to make me grow new teeth or something to get me to give Nvidia money over Intel at this point.
You kinda missed the most important detail: they’re competing with the mid-range (and yes, a 4060 is the midrange) for substantially less money than the competition wants.
I know game nerd types don’t care about that, but if you’re trying to build a $500 gaming system, Intel just dropped the most compelling gpu on the market and, yes, while there’s an upcoming generation, the 60-series cards don’t come out immediately, and when they do, I doubt they’re going to be competing on price.
Intel really does have a six month to a year window here to buy market share with a sufficiently performant, properly priced, and by all accounts good product.
I said that in my comment. And no, 4060 is not midrange lol
4090 48GB
4090
4080 Super
4080
4070 Ti Super
4070 Ti
4070 Super
4070
4060 Ti 16GB
4060 Ti
4060
It’s literally the lowest end GPU they make. The 60-class GPU stopped being midrange for Nvidia with Pascal, although due to Nvidia’s exceptional marketing capability, they’ve tricked people into thinking that’s not the case.
My main complaint isn’t with the performance, but the missed opportunity to release a higher SKU with more RAM. 12GB is enough for gaming with their performance, but adding more would open up other uses, like AI or other forms of compute. Maybe they still will, idk, but I would be totally willing to upgrade my AMD GPU if there was a compelling reason beyond a little better performance. Give me 16 or even 24GB VRAM for $300 or so and I’d buy, even if it’s not “ready” at launch (i.e. software support for AI/compute).
As of now, the GPU is well placed for budget rigs, but I think they could’ve cast their net a bit wider.
And that’s why I said it’s well placed for budget rigs. If I was building a computer today, I’d probably go with the B580.
However, I already have a computer with an RX 6650 XT, and while the B580 is an upgrade (10-15% higher FPS, esp at higher resolutions), it’s not enough to really convince me to upgrade. However, a higher RAM variant would because it adds capabilities that I can’t get with my current card.
Intel needs marketshare, and a high VRAM SKU would get a lot of people talking. They don’t even need to sell a lot of that SKU to make a big difference, it just needs to exist and have decent software support. They could follow it up with an enterprise lineup targeted at AI and GPGPU once the SW ecosystem is solid (which enthusiasts like me will help test).
Sounds like sour grapes, which is pretty much the only thing of note coming out of Intel lately
This sounds pretty plausible. The windows user is the least likely to understand the implications of arm for their applications in the ecosystem that is the least likely to accommodate any change. Microsoft likes to hedge their bets but generally does not have a reason to prefer arm over x86, their revenue opportunity is the same either way. Application vendors not particularly motivated yet because there’s low market share and no reason to expect windows on x86 to go anywhere.
Just like last time around, windows and x86 are inextricably tied together. Windows is built on decades of backwards compatibility in a closed source world and ARM is anathema to x86 windows application compatibility.
Apple forced processor architecture changes because they wanted them, but Microsoft doesn’t have the motive.
This has next to nothing to do with the technical qualities of the processor, but it’s just such a crappy ecosystem to try to break into on its own terms.
Ironically, their new GPUs are supposedly actually pretty decent. It’s like Bizarro-world over there, LOL!
Their CPUs would also be decent if they only made low end parts
Burn unit to Intel!
Decent only if you look at raw performance for the price compared to other MSRPs.
When you scratch beneath the surface a little and see what they’re having to do to keep up with the 3 year old low end Nvidia and AMD parts (that are due to be replaced very soon), it paints a less rosy picture. They’re on a newer, more expensive node, use a fair bit more power, and have a larger die size by quite a bit than their AMD/Nvidia counterparts.
Add to that Intel doesn’t get the discounts from TSMC that Nvidia and AMD get, and I’m doubtful Battlemage is profitable for Intel (this potentially explains why availability has been so poor - they don’t want to sell too many).
While it’s true the average buyer won’t care about the bulk of that, it does mean Intel is limited in what they can do when Nvidia and AMD release their next generation of stuff within the next few months.
If I can get one I’m buying one. I think their performance/cost ratio is excellent, and will probably make NVidia and AMD bring down their mid-range card prices.
But I’m not forgetting who made the prices come down. I’m all in on supporting a new player in the GPU game, and the 5060 would have to make me grow new teeth or something to get me to give Nvidia money over Intel at this point.
You kinda missed the most important detail: they’re competing with the mid-range (and yes, a 4060 is the midrange) for substantially less money than the competition wants.
I know game nerd types don’t care about that, but if you’re trying to build a $500 gaming system, Intel just dropped the most compelling gpu on the market and, yes, while there’s an upcoming generation, the 60-series cards don’t come out immediately, and when they do, I doubt they’re going to be competing on price.
Intel really does have a six month to a year window here to buy market share with a sufficiently performant, properly priced, and by all accounts good product.
I’m sorry that the facts surrounding Nvidia’s GPUs upset you.
I said that in my comment. And no, 4060 is not midrange lol
4090 48GB
4090
4080 Super
4080
4070 Ti Super
4070 Ti
4070 Super
4070
4060 Ti 16GB
4060 Ti
4060
It’s literally the lowest end GPU they make. The 60-class GPU stopped being midrange for Nvidia with Pascal, although due to Nvidia’s exceptional marketing capability, they’ve tricked people into thinking that’s not the case.
My main complaint isn’t with the performance, but the missed opportunity to release a higher SKU with more RAM. 12GB is enough for gaming with their performance, but adding more would open up other uses, like AI or other forms of compute. Maybe they still will, idk, but I would be totally willing to upgrade my AMD GPU if there was a compelling reason beyond a little better performance. Give me 16 or even 24GB VRAM for $300 or so and I’d buy, even if it’s not “ready” at launch (i.e. software support for AI/compute).
As of now, the GPU is well placed for budget rigs, but I think they could’ve cast their net a bit wider.
Yeah, you’re an enthusiast looking for enthusiast parts.
Try to understand that you’re not the only people in the market or discussion.
And that’s why I said it’s well placed for budget rigs. If I was building a computer today, I’d probably go with the B580.
However, I already have a computer with an RX 6650 XT, and while the B580 is an upgrade (10-15% higher FPS, esp at higher resolutions), it’s not enough to really convince me to upgrade. However, a higher RAM variant would because it adds capabilities that I can’t get with my current card.
Intel needs marketshare, and a high VRAM SKU would get a lot of people talking. They don’t even need to sell a lot of that SKU to make a big difference, it just needs to exist and have decent software support. They could follow it up with an enterprise lineup targeted at AI and GPGPU once the SW ecosystem is solid (which enthusiasts like me will help test).
Tell me about it