Why does AMD suck?

[pelov]

I'm not sure "suck" is the right word.

The 6300 and 4300 are both pretty good chips, especially at their respective prices. With unlocked multipliers, those chips remind me of the old Denebs as far as performance and "bang for your buck." They'll slap around the core i3's in pretty much every statistic other than perf-per-watt and power consumption.

The 8350 and 8320 are decent productivity chips, though that's a much tougher sell as these people don't mind spending the extra money on a better performing CPU (see 3770K or 2011 platform).

AMD "sucks" because they have a small R&D budget and they've had to put up with a weak GloFo. Bulldozer really screwed them over, though. Had they released Vishera last year I think the response from the enthusiast community and their investors would have been much more receptive rather than dark and gloomy.

 

[cytg111]

Phenom II is a decent chip and aged better than Core2Quad.

The problem is it arrived when Nehalem was out.

Didnt the P2 pack ~15% over P1?

Much like Vishera ...

 

[Yuriman]

Didnt the P2 pack ~15% over P1?

Much like Vishera ...

A lot more than that once you factored in overclocking. PhII was a bigger jump relative to Ph1 than V vs BD.

 

[moonbogg]

They won't suck for long in the PC market. They are leaving it. The only way to catch up would be for Intel to make a SERIES of big, catastrophic mistakes.

 

[inf64]

Where Vishera improves greatly over BD is games (and I mean "IPC"). Those small improvements all over the place inside the core add up in branchy code that games have so Vishera is now much better chip than BD was and clearly superior to X4/X6.

 

[mrmt]

Phenom II is a decent chip and aged better than Core2Quad.

The problem is it arrived when Nehalem was out.

AMD 486 were also descent too, the problem was that it arrived when Pentium was out.

Except when Intel was disoriented making netburst, AMD is at least one generation behind Intel. Short story, AMD is a day late, a dollar short.

 

[CHADBOGA]

On topic of sucky AMD chips,one poster at XS noticed that 8350 is already sold out at the newegg . I didn't believe it so I checked it and sure,it's already gone . For a sucky CPU it just sold in around 10K units which is supposedly (what the guy at XS says) a number of units Newegg gets.

Or maybe they just got a couple of hundred units and sold them easily...

We had this same thing with Bulldozer.

Do you think Bulldozer was an outstanding success?

 

[Idontcare]

On topic of sucky AMD chips,one poster at XS noticed that 8350 is already sold out at the newegg . I didn't believe it so I checked it and sure,it's already gone . For a sucky CPU it just sold in around 10K units which is supposedly (what the guy at XS says) a number of units Newegg gets.

Or maybe they just got a couple of hundred units and sold them easily...

Bummer, but see this just goes to show that Newegg knows what it is doing when it raises its prices to take supply/demand into account.

People whine that newegg is gouging, but sold-out means no one gets one at any price. So why not raise price to match demand to supply?

But I was hoping for Neweggs prices to go the other direction. Maybe by cybermonday with more supply in hand?

 

[NTMBK]

I find it very odd that this thread is still open, but the equally trolling "why does intel suck" thread got closed, and piesquared is banned but this thead's OP isn't.

 

[gammaray]

AMD sucks because for the past 10 years they reported losses in each of the freaking quarters except 2.

still amazing they're still in business.

But yeah, once AMD disappears from the cpu markets, it will suck for us consumers.

 

[ShintaiDK]

AMD sucks because for the past 10 years they reported losses in each of the freaking quarters except 2.

still amazing they're still in business.

But yeah, once AMD disappears from the cpu markets, it will suck for us consumers.

Why? One could say that, in the long term development in the semiconductor business, competition is directly harmful due to ROI.

Competition is far from the holy grail that some think. Competition can be great, but it can also be harmful. For example when companies simply cut costs for a product to be barely functional and reliable in the hunt of cost savings in competition. Or ends up with products thats directly harmful for you as seen with food.

 

[zephxiii]

Because AMD keeps Intel from lettings SKUs slowly trickle out and milk the hell out of us.

 

[Olikan]

Competition is far from the holy grail that some think. Competition can be great, but it can also be harmful. For example when companies simply cut costs for a product to be barely functional and reliable in the hunt of cost savings in competition. Or ends up with products thats directly harmful for you as seen with food.

IMO,
it's good when all companies have a decent market share, so they can cover the R&D from new and better products...

it sucks when there is like intel and AMD...one company don't have R&D to fight back, while the other can just stay still and laugh

 

[ShintaiDK]

Because AMD keeps Intel from lettings SKUs slowly trickle out and milk the hell out of us.

No, Intel cant raise prices since volume then drops. Lower volume means lower fab utilization, higher node cost per chip and lower margins. This again means lower revenue and lower profit. Intel needs higher volume to offset the ever increasing cost of fabs and designs.

Same with innovation and why Intel have been its own sole competitor since Core 2.

 

[ShintaiDK]

IMO,
it's good when all companies have a decent market share, so they can cover the R&D from new and better products...

it sucks when there is like intel and AMD...one company don't have R&D to fight back, while the other can just stay still and laugh

Imagine 5 AMDs today. Do you think that would be good for you and me or the future?

 

[Olikan]

Imagine 5 AMDs today. Do you think that would be good for you and me or the future?

LOL, why it had to be 5 AMDs? can't be 5 intels?

 

[ShintaiDK]

LOL, why it had to be 5 AMDs? can't be 5 intels?

5 Intels would require a volume and revenue 300% higher than the entire x86 market is today.

 

[Olikan]

5 Intels would require a volume and revenue 300% higher than the entire x86 market is today.

...i think that you didn't read between the lines...

anyway, i do think that 5 companies sharing all the revenue of x86 market
would be way better

 

[ShintaiDK]

...i think that you didn't read between the lines...

anyway, i do think that 5 companies sharing all the revenue of x86 market
would be way better

If that was so, we would most likely cheer over Core 2 and Phenom performance today on 45nm, assuming the companies needed to run with a small profit.

It certainly wouldnt be better. It would be a direct disaster.

 

[zephxiii]

No, Intel cant raise prices since volume then drops. Lower volume means lower fab utilization, higher node cost per chip and lower margins. This again means lower revenue and lower profit. Intel needs higher volume to offset the ever increasing cost of fabs and designs.

Same with innovation and why Intel have been its own sole competitor since Core 2.

No, but they can slow down the release of newer/faster chips...

 

[nehalem256]

AMD 486 were also descent too, the problem was that it arrived when Pentium was out.

Except when Intel was disoriented making netburst, AMD is at least one generation behind Intel. Short story, AMD is a day late, a dollar short.

Well and the original Athlon was a solid competitor or superior to the P3. This was the situation for at least 1 year before the P4 came out.

I wonder if a lot of the problem is it seems like very other process node AMD uses sucks. And unfortunately for AMD they got stuck with their sucky 65nm process when the original Core 2 came out. And are stuck with their sucky 32nm process today. So hopefully the 28nm SR works out better

 

[ShintaiDK]

No, but they can slow down the release of newer/faster chips...

No, lack of innovation leads to people delaying upgrading and then volume drops. And you are back again in losing as a company.

 

[zephxiii]

Well you aren't going to convince anyone that a monopoly is best lol. I will take market competition over that anyday. I do agree that there can't bee too many players, but we need at least one other.

 

[ShintaiDK]

Well and the original Athlon was a solid competitor or superior to the P3. This was the situation for at least 1 year before the P4 came out.

I wonder if a lot of the problem is it seems like very other process node AMD uses sucks. And unfortunately for AMD they got stuck with their sucky 65nm process when the original Core 2 came out. And are stuck with their sucky 32nm process today. So hopefully the 28nm SR works out better

One reason is also the ROI involved. It was simply much cheaper back then. Meaning it was easier to compete.

 

[ShintaiDK]

Well you aren't going to convince anyone that a monopoly is best lol. I will take market competition over that anyday. I do agree that there can't bee too many players, but we need at least one other.

Competition aint a grant saviour. Competition is in some cases directly damaging. Either because the quality, service and/or reliability drops. Or because the individual R&D budgets gets too low.

Competition can also damage innovation, since the rick factor compared to staus quo can be too high. This is for example why governments subsidizes or directly invest into new business areas, because the regular market competition is against the step due to self interest.