Anybody Remember Conroe?

[Arachnotronic]

All right, so I don't know how many of you are old-timers, but now that we're seeing the first performance leaks of "Bay Trail" (following, of course, projections from Intel), does anybody else get the same sort of vibe that was around when Conroe was about to come upon us?

Right now, we have a few folks that believe that Intel's finally going to "do it" and take leadership in this space, but the many folks seem to be overly critical/dismissive of Intel's efforts, offering up arguments similar to the following,

- That benchmark doesn't count [EDIT: In the case of AnTuTu, which really is the only benchmark we have right now, Exophase makes a very compelling case for why it really doesn't count which is well worth reading below ]
- ARM has something better coming at 20nm, you just wait and see (even though TSMC doesn't expect 20nm to even be 2% of revenues until 2Q 2014)

Steven C. Pelayo - HSBC, Research Division
What quarter will be the first few percentage of revenues will come from 20-nanometer for TSMC?

Morris Chang - Chairman and Chief Executive Officer
What quarter will be the first 2% quarter?

Lora Ho - Chief Financial Officer and Senior Vice President of Finance
Well, based on our current estimation, it will be roughly second quarter 2014.

- Oh, but will Intel price it too high? (even though if Intel doesn't, it simply won't sell any, which would be deadly as the traditional PC upgrade cycles lengthen and sales in that space fall off)

- Intel's graphics suck!

It just seems to me the same kind of dismissal that happened when Intel introduced Conroe ("Oh, Barcelona will be better"), and then Kentsfield ("It's not true quad core!"), and then Nehalem, and then Sandy Bridge ("BULLDOZER WILL DESTROY IT!"), etc.

While I don't expect the likes of Qualcomm to end up the way of AMD, I'm pretty stunned at the near-universal denial/hatred of Intel that I'm seeing here. We know that the *only* reason that Intel hasn't been competitive in this space is because they've been fighting with a POS 5 year old core. Everything else is in place for tablets...Windows 8 is much better than Windows RT, Bay Trail looks to be a great performer at low power consumption, and Intel has a process advantage that's eerily similar to the kind of advantage Intel has had over AMD all of these years.

And on top of all that, when Intel actually responds to what the market wants with Haswell (battery life, battery life, and battery life), people [scoff at] it, claiming that Intel is "being lazy". I'm sorry, but Sandy Bridge -> Haswell is a pretty good jump, and the reason IVB -> Haswell didn't look as impressive is that IVB, unlike most "ticks", actually brought a 3-5% IPC increase.

I don't know...I just don't get all this hatred for Intel around here, although I suppose nobody ever really cheers for Goliath.

Thoughts?

Profanity is not allowed in the technical forums. I'm also not happy about the tone of this post, though it doesn't necessarily violate our rules
-ViRGE

 

[Cerb]

Thoughts?

Reads like a baiting troll post, to me.

<- posting from one of several Conroe-powered computers

In the future please avoid these kinds of post. They are thread crapping
-ViRGE

 

[Arachnotronic]

Reads like a baiting troll post, to me.

<- posting from one of several Conroe-powered computers

I sincerely hope that this isn't coming off as a "troll" post...I'm honestly just interested in understanding why there's so much hate/doubt on the Intel side of things. I know there are some totally rational posters on this forum who take the opposite view that I do, and I really look forward to reading what they have to say.

If the moderators think it's a troll post, I will be happy to modify it to their wishes.

 

[Sweepr]

If theres something Conroe showed is to never underestimate Intel. Haswell was great for mobile, outstanding battery life (up to 60% more than IB under light workloads), solid gfx performance improvement (I would have laughed if anyone told me that two years from Llano vs SB Intel would have the fastest IGP aka GT3e) and a respectable ~10% IPC bump from IB (which is 3-5% faster than SB @ equal clocks). Bay Trail is based on brand new cores, thats why expectations are so high.

 

[Idontcare]

Reads like a baiting troll post, to me.

<- posting from one of several Conroe-powered computers

We seriously need to have less of this kind of posting around here. Be it an Intel thread, an AMD thread, an ARM thread, an APU thread, an all rounder thread, a steamroller thread, etc, etc, etc...

Posting a pre-emptive thread-crap post that derails a thread before it gets off the ground is just not productive.

Sorry to pick on your post Cerb, its not that you did a bad thing, what you did here is commonplace in the CPU forum...but it is also a destructive thing to do and we need less destructive stuff and more productive stuff.

If a thread sets off your spidey-sense and makes you all tingly thinking it is a bait or troll thread then please just report it to the mods and refrain from posting publicly within the thread itself.

If the moderators think it's a troll post, I will be happy to modify it to their wishes.

This IS the place for this very sort of "lets talk about this" type threads. If the community finds it interesting then they will join the conversation and the thread will stay relevant and on top. If the community isn't interested then it will die a quick and silent death as it plummets off the front page.

 

[Arachnotronic]

Idontcare,

Absolutely

 

[blackened23]

Well...of course you're going to see that kind of thing, it happens on nearly every forum. I won't condemn what they (AMD / ARM / etc fans) are saying, everyone picks their favorites I suppose. With that being said, I personally think intel will achieve dominance in the mobile space because:

1) Intel has a dominating technological and performance lead
2) Intel was never focused on efficiency and mobility until the past two years
3) Intel is overflowing with cash for R+D
4) Their new CEO is completely focused and hellbent on dominating mobile

Cash and R+D translate into dominance in the silicon industry, i'm sorry to say the old days of the little guys being able to stay relevant are long, long gone. I mean, I appreciate AMD for what they are - I remain a big fan of their GPUs - but I just can't see them becoming a major threat in the desktop / mobile space anytime soon just because they don't have the cash to invest into R+D, while intel does. As much as some like to trash-talk AMD, AMD *does* have some fantastic engineers. Some of the best in the business, even - but that just hasn't translated so far because their resources are very limited. So now that intel has focused on the task at hand - I think they will win. Maybe some will interpret that as me being a fanboy. Well, I believe intel will do it because they have a history of delivering. When intel states that they're completely focused on delivering their highest performance and efficiency for mobile products, I tend to take it at face value because they have been very consistent on delivering on their promises.

Hopefully this isn't inappropriate to state, as this may be cynical - but I guess I want to see ARM Holdings get very uncomfortable. Some of the statements they have made over the prior 2 years with regard to Intel have been completely arrogant, such as "intel will never beat us in efficiency". It's just funny to see hubris catch up to a company, you know? So I suppose i'm rooting for Intel because of this reason. Call me a fanboy if you wish.

 

[USER8000]

The problem is the race to the bottom and ultimately technical superority is not necessarily an indication of 100% guaranteed success. There are many other factors at play like cost or IP licensing terms for example. The problem is people are trying to take experience with desktops,etc and extrapolate from that.

People have forgotten how ARM was written off decades ago too for use in personal computing,and never were expected to be ANY competition in the personal computing field to big companies like Intel. People are also forgetting the financial resources companies like Qualcomm,Apple,Samsung have which are far more than what AMD could ever muster even at their height and probably can challenge or exceed what Intel has. They can probably even afford to make mistakes along the way(unlike AMD). That is who Intel are fighting,meaning everyone from a small Chinese developer with razor thin margins to giants like Qualcomm.

If those companies pull their full weight behind developing ARM CPUs,it will be just an all out fight,and they are not going to keel over and say meh.

Then there are the Chinese developing their own chips too using the MIPS instruction set as a national product with government backing,which might end up starting to penetrate more markets.

So whether Intel wins,Qualcomm wins,or everyone wins is all uncertain IMHO.

In the end I only care that I get better final products at a reasonable price. Hopefully a good old fight between these giant tech firms will ensure this.

 

[blackened23]

The thing about ARM SOCs is that they are typically very low performance yet high efficiency devices. ARM Holdings essentially became relevant because the goalposts shifted from achieving the highest possible performance/IPC to the highest possible efficiency for on-the-go devices. That isn't something that Intel was necessarily focused on in prior years, but they now are completely focused on that.

 

[wlee15]

As Exophase explained Antutu is complied with ICC which allows Silvertmont to execute vector SSE instructions while the ARM side is complied to support only the scalar VFP instructions.

 

[Idontcare]

We all win if Intel Conroe's the mobile market. We all lose if Intel is the last company standing though.

So I hope Intel rocks the mobility world and compels these companies with juicy fat gross margins to take a lower margin and boost their R&D budgets to go toe-to-toe with Intel.

I'm ok with these companies spending every last dollar of their profit on expanding their R&D efforts as they duke it out.

If AMD hadn't Athlon'ed the market then Intel would not have been compelled to Conroe it. Unfortunately for all of us Ruiz wasted AMD's opportunity such that AMD was in no shape to weather a Conroe-sized disruption. Qualcomm, Apple, and Samsung are in great shape (too good of shape if you ask me, does Apple really need $145B cash on hand? that's a lot of potential R&D!).

And I do agree with the OP in the spirit of the opening post which is basically a way of saying that nothing seems to draw out the haters more than a company (any company) doing well for itself. But that is only natural, humans are wired to be jealous and spiteful, we tear down people who are successful and we do the same with institutions, governments, universities, and businesses.

It is our nature to loathe the success of others if it is not directly benefiting us in some way. Ain't nothing going to change that

 

[Exophase]

My criticism of AnTuTu is not denial.

Your criticism of my criticism, without actually providing any technical refutation of what I said, now that's denial.

Silvermont could be the best thing ever and could absolutely destroy anything any official or custom ARM core ever can or ever will do, but you'll never see it with AnTuTu because it's a broken benchmark.

Unfortunately I currently can't give more details than I did in the comment in your SA article, but these are easily verifiable for anyone who has the smallest familiarity with how to use an assembler and how to interpreter x86 and ARM ASM. Which is a lot to ask, I know.

 

[jaqie]

I found the original post to be loaded to the gills with the sort of thing saying "anyone that denies this is a fanboy of anything else" but on the original point of the article I will comment.

I personally go with what is best bang for the buck for my budget at the time - this includes some things many people don't usually consider, that I don't care to delineate right now.

Suffice it to say that I went from athlon64 toledo x2 3800+ to a conroe c2d @3GHz

rehosted at imageshack, sorry. fixed.

and after that I went to a q6600@3GHz and then a phenom II thuban 6 core @ 3.5GHz. If I had money right now I would get one of the ivy bridge or haswell i7, and would love them.

This new stuff coming out of intel is just insanely good, honestly. It feels like the opposite of when the pentium 4 was out and amd was mopping the floor with intel. I really do hope things swing around the other way again for the sake of competition making both companies healthy, but right now I'll say again that intel has the stuff.

And this is coming from someone that just got an HP E-300 laptop for $270 new at walmart... yeah, if I had the money this would be an i7 haswell laptop, and I would be so much happier.

 

[frozentundra123456]

I sincerely hope that this isn't coming off as a "troll" post...I'm honestly just interested in understanding why there's so much hate/doubt on the Intel side of things. I know there are some totally rational posters on this forum who take the opposite view that I do, and I really look forward to reading what they have to say.

If the moderators think it's a troll post, I will be happy to modify it to their wishes.

My first computer was a pentium 233 mhz, so yes I am that old, and yes, I remember conroe. TBH, I dont see that kind of quantum improvement coming ever again. I dont really follow the low power segment, but no matter how good silvermount is, ARM is already entrenched in that market, so Intel will have a difficult time breaking into it, and I think their margins will suffer.

That said, I do agree with you about the general climate of the forums lately. I think I posted already in the other thread about this, but it seems that every powerpoint slide or undocumented claim from a developer or company rep is taken as gospel from one company, while any positive data about their competitor is discounted as cheating(the software is biased, yada, yada), incorrect or does not matter because of excuse x,y,or z. Even more ironic is that anyone who refutes the unfounded, outrageous claims is accused of hating on their favorite company.

 

[Maximilian]

All right, so I don't know how many of you are old-timers, but now that we're seeing the first performance leaks of "Bay Trail" (following, of course, projections from Intel), does anybody else get the same sort of vibe that was around when Conroe was about to come upon us?

Old timers? Conroe?

Are you kidding me

 

[Arachnotronic]

My criticism of AnTuTu is not denial.

Your criticism of my criticism, without actually providing any technical refutation of what I said, now that's denial.

Silvermont could be the best thing ever and could absolutely destroy anything any official or custom ARM core ever can or ever will do, but you'll never see it with AnTuTu because it's a broken benchmark.

Unfortunately I currently can't give more details than I did in the comment in your SA article, but these are easily verifiable for anyone who has the smallest familiarity with how to use an assembler and how to interpreter x86 and ARM ASM. Which is a lot to ask, I know.

I don't doubt that AnTuTu isn't exactly the best benchmark; I agree, it gives whacky results. I don't believe for a second that the FPU performance of a 2C/4T CT+ could come anywhere close to a quaddie Krait 300 or Cortex A15.

But my point is: the numbers given at the Silvermont launch (which I believe were mostly SPECint_rate_2000 in the comparison charts) were certainly enough to get me (and I'm sure others) to at least think that Intel will have a fighting chance in this space. The AnTuTu bench, while not perfect, shows that generation -> generation (that is Z2580 -> Z3770) the performance gains were in-line with what was promised.

Like I said...do I expect the ARM guys to roll over and die? Hell no. But do I expect Intel to be a viable player in this space, and do I look forward to some nice Windows 8 tablets? Very much so...the current crop of 1366x768 CT tablets really aren't enough to fight in this space and aren't anything I'd plunk down $400 for, especially given what the iPad and Nexus 10 currently offer.

 

[Exophase]

I don't doubt that AnTuTu isn't exactly the best benchmark; I agree, it's weird.

But my point is: the numbers given at the Silvermont launch (which I believe were mostly SPECint_rate_2000) were certainly enough to get me (and I'm sure others) to at least think that Intel will have a fighting chance in this space.

It's not "not exactly the best benchmark" - every mobile benchmark I've seen sucks with the possible exception of the 3D ones. It's more like it's the worst one I'm currently aware of, although I haven't looked into others to the same extent so they could be hiding out there.

But knowing that doesn't stop you and others from getting excited and bragging about AnTuTu leaks, or posting your own AnTuTu scores. Let's get this straight, when I said the benchmark doesn't count it's because it doesn't count. And yet you have the nerve to say it's because I'm in denial.

AnTuTu isn't even reliable for comparing Saltwell and Silvermont if you don't first make sure it's the same version. Did you make sure it's the same version of AnTuTu? I can't explain right now exactly why that matters but use your imagination. Even if it WAS the same version, scaling nicely in a broken and/or pointless benchmark is meaningless. Would you use SuperPi this way?

If you guys are really interested in Silvermont you should check out the Intel optimization guide which was just updated to include information about it. The funny thing is that it's not nearly as different from Saltwell as a lot of people are saying it is (I also always thought Saltwell wasn't as bad as people were saying, but that's something else). It's still two-wide, yes, but the SIMD/FP and memory pipelines are both still in-order, and the SIMD/FP execution units and latency are pretty much identical to Saltwell. The memory pipeline is also still only load or store at a time. Fetch and decode is pretty much the same aside from fixing the x87 screwup which doesn't really matter. At a high level it looks an awful lot like the difference between Cortex-A8 and A9 - the big improvements are the OoOE dual integer pipes with unified scheduler and decoupled int/mem/FP pipes allowing out of order completion and reducing the branch mispredict penalty.

That doesn't mean that it can't still embarrass everyone else (for one thing by clocking higher for the same power consumption) and of course it's all a balace in terms of perf/W - the devil's in the details and a lot of these details aren't immediately apprarent. But people are acting like Intel started totally from scratch with everything when the reality is closer to a core revision.

 

[Hulk]

We all win if Intel Conroe's the mobile market. We all lose if Intel is the last company standing though.

So I hope Intel rocks the mobility world and compels these companies with juicy fat gross margins to take a lower margin and boost their R&D budgets to go toe-to-toe with Intel.

I'm ok with these companies spending every last dollar of their profit on expanding their R&D efforts as they duke it out.

If AMD hadn't Athlon'ed the market then Intel would not have been compelled to Conroe it. Unfortunately for all of us Ruiz wasted AMD's opportunity such that AMD was in no shape to weather a Conroe-sized disruption. Qualcomm, Apple, and Samsung are in great shape (too good of shape if you ask me, does Apple really need $145B cash on hand? that's a lot of potential R&D!).

And I do agree with the OP in the spirit of the opening post which is basically a way of saying that nothing seems to draw out the haters more than a company (any company) doing well for itself. But that is only natural, humans are wired to be jealous and spiteful, we tear down people who are successful and we do the same with institutions, governments, universities, and businesses.

It is our nature to loathe the success of others if it is not directly benefiting us in some way. Ain't nothing going to change that

It would be great it Intel "Conroe'd" the mobile market as the ARM guys are well situated cash-wise to fight. We *could* see unbelievable competition and resulting leap frogging for a couple of processor generations which would be amazing.

 

[Cerb]

I sincerely hope that this isn't coming off as a "troll" post...

Maybe I'm jaded with this subforum .

I'm honestly just interested in understanding why there's so much hate/doubt on the Intel side of things.

Intel is probably having to spend as much as ARM and their licensees combined to get where they're going, if not more (they do spend more; I mean spending just to improve cost and Wh-efficiency). They're huge. In general, people want David to defeat Goliath, not Goliath to starve David. Intel holds that Goliath position. Due to their ability to spend, they will succeed. It's just a question of how soon, and at what cost. ARM has historical expertise, but Intel can spend enough to get their engineers up to speed at pretty much the fastest rate that it can be done, and the most interest ARM chips are just ISA-licensed.

Haswell offers 30%+ battery life, which is pretty impressive. The new IGP isn't bad, and Intel is caring about driver quality, finally. We'll see soon how well Silvermont actually stands up.

But, neither ARM nor Intel have any pixie dust. Intel is spending more and more money. It can't go on forever, even with their typical margins. Mobile ARM companies, and probably near-future makers of ARM server SoCs, are awfully conservative in their spending, but do well serving markets that aren't affected too much by performance improvements, as long as they're there (IE, as long as gen n+1 solidly beats gen n, making n+1 20% another faster would be wasted, or R&D funds that could go into making them cheaper, in high volume, instead).

 

[Arachnotronic]

But knowing that doesn't stop you and others from getting excited and bragging about AnTuTu leaks, or posting your own AnTuTu scores. Let's get this straight, when I said the benchmark doesn't count it's because it doesn't count. And yet you have the nerve to say it's because I'm in denial.

I wasn't referring to you at all. I am referring to the *countless* media pundits who see graphs such as these:

And then say that Intel still won't be competitive, or that the benchmarks listed may be compiled for Intel only, or that the battery life won't be good, it'll cost too much, etc... Of course we should ALWAYS take what companies say with a grain of salt, but if these are even *remotely* true, then...this really is a "Conroe'ing".

If they're not true, Intel will look like complete and utter fools and the company will slowly but surely decay as other, more aggressive firms take more of the computing market, and Intel's R&D + capex can no longer be funded, further exacerbating said death-spiral. Which, I mean, they totally did when they were stumbling over themselves with the Pentium 4...fixed it a bit with Northwood, then screwed it all up with PresHott, but I don't really get that impression.

 

[Arachnotronic]

If you guys are really interested in Silvermont you should check out the Intel optimization guide which was just updated to include information about it. The funny thing is that it's not nearly as different from Saltwell as a lot of people are saying it is (I also always thought Saltwell wasn't as bad as people were saying, but that's something else). It's still two-wide, yes, but the SIMD/FP and memory pipelines are both still in-order, and the SIMD/FP execution units and latency are pretty much identical to Saltwell. The memory pipeline is also still only load or store at a time. Fetch and decode is pretty much the same aside from fixing the x87 screwup which doesn't really matter. At a high level it looks an awful lot like the difference between Cortex-A8 and A9 - the big improvements are the OoOE dual integer pipes with unified scheduler and decoupled int/mem/FP pipes allowing out of order completion and reducing the branch mispredict penalty

I read through it this morning. I, too, was pretty surprised at how...not aggressive the FPU seems to be. It's funny though, as I actually asked one of the Atom architects about this. Here was the individual's response...take it for what it's worth:

1. Could you please help me understand the competitive positioning of Silvermont to A15 and, perhaps, A57? Intel claimed that Silvermont IPC will be about on-par with A15, despite having a narrower front end/back end, smaller OoO structures, lack of FMAC capability, etc. So, my real question is: what gives you confidence that Silvermont stays this competitive with A15 on integer workloads (this is what the competitive comparison at the uArch disclosure talked about) despite the narrower design? Further, should we expect that Silvermont is as competitive on floating point with the ARM design as it is on integer (this is what was highlighted at the talk)?



We have A15 measurements, and Silvermont measurements, so confidence here is based on actual measurements we’ve seen. (We cross-check these results with what ARM quotes themselves)

We’ve done a lot of analysis and there are many areas where we have an advantage: our memory latency (at each level of cache hierarchy) and memory BW, our superior branch handling, our complex-instruction support, our smart resources management, and many other things, allow us to outperform.

From Silverthorne to Silvermont, gains on FP workload should be even better than gains on integer workloads. Again, we have A15 measurements, and Silvermont measurements, and are confident in our FP scores are similarly impressive.

 

[Arachnotronic]

I'm kind of puzzled here. Mods are continuously telling us to remain technical but OP just spelt his gut out cheering Intel with ZERO technical reasons. He is refuting arguments with nothing but more cheerleading. He even has an "Intel" tattooed on his nickname for TFSM' sake.

I'm with Cerb here, this is a baiting thread. For me it's pretty clear what "Conroeing" the mobile market means.

I'm seriously disappointed and IDC encouraging this doesn't make it any better.

I'll be honest with you - I made this screen-name many years ago (2006?) and for lack of anything better to choose for a tech forum, I just used the brand of the CPU I had at the time and a number that I liked.

Could've very easily been Nvidia17 or Corsair17

Anyway, look at the post above you...just provided an interesting tid-bit from a dialog with one of the folks on the Atom team. Could be worth discussing.

 

[sniffin]

I love and despise Intel. I think their engineers are ridiculously talented and have immense respect for that part of the company. What I don't like is the non technical side of the company. The same side that are prepared to make their products worse for the slimmest reductions in cost.

I get that companies are interested in maximizing profit. What I don't enjoy is companies pursuing profit so aggressively they are prepared to completely ignore a small but enthusiastic part of their customer base. Do you really think Intel soldering the IHS to the die for their 'k' CPUs would have had any real impact on their bottom line?

But what annoys me even more is when people defend these decisions. You reference the fact some people are fans of other companies and your disdain (I get that). People call out AMD fanboys all the time. And yet these same people will turn around and brush aside the deliberate crippling of Intel's products and cite the marginal performance increases we've observed for the last two generations. These are the people Intel couldn't give [a hoot] about. And they have the hide to mock people who defend AMD.

No profanity in the tech forums
-ViRGE

 

[beginner99]

I wasn't referring to you at all. I am referring to the *countless* media pundits who see graphs such as these:

And then say that Intel still won't be competitive, or that the benchmarks listed may be compiled for Intel only, or that the battery life won't be good, it'll cost too much, etc... Of course we should ALWAYS take what companies say with a grain of salt, but if these are even *remotely* true, then...this really is a "Conroe'ing".

If they're not true, Intel will look like complete and utter fools and the company will slowly but surely decay as other, more aggressive firms take more of the computing market, and Intel's R&D + capex can no longer be funded, further exacerbating said death-spiral. Which, I mean, they totally did when they were stumbling over themselves with the Pentium 4...fixed it a bit with Northwood, then screwed it all up with PresHott, but I don't really get that impression.

Exactly. even if you adjust the slides and say performance is only 1.5x better and power is only 2x lower it's still pretty good. I mean if it performs even worse than those "halved numbers", it's a lot bigger Marketing failure than JF-AMD with BD.

 

[tential]

I think this will be like Conroe for the mobile market. Either that, or Conroe for the mobile market is coming VERY soon. We saw what happened when intel looked like it was having trouble in the Desktop realm. They laid the hammer down, and with such force that AMD still looks like a scared puppy in a corner, unable to ever catch up really desktop side.

In all honesty, I hope intel doesn't dominate this market as well. We need competition, and a "Conroe" might be something that fights that.