AMD CPU and Platform Discussion: Ryzen Might Be...Good? *knocks on wood*

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > AMD CPU and Platform Discussion: Ryzen Might Be...Good? *knocks on wood*

«‹›3 »

WhyteRyce: Dec 30, 2001

Alereon posted:

I happened to be looking at Zacates on Newegg, and saw this new Sapphire option that looks good:

SAPPHIRE Pure White Fusion E350 for $109.99

It has a Marvell Yukon Ethernet chipset which tend to work better than Realtek for some people. I'm just bummed we don't have more options with passive heatsinks, you have to spend $145 for the Asus, though it is a nice product with extra eSATA and USB3.0.

All the mITX boards seem to come with some noisy rear end high pitched fan. The MSI board I have has a tiny fan that spins at 5k+ RPM with limited options to quiet it. I ended up just sticking resistor on it to shut it up.

I was looking at getting the Asus mITX board with its giant passive cooling heatsink but the reviews on Newegg made it sound like some people had heat issues

# ¿ Apr 22, 2011 19:03

Adbot: ADBOT LOVES YOU

# ¿ May 4, 2024 06:21

WhyteRyce: Dec 30, 2001

Alereon posted:

For an unoverclocked system, you definitely want 1866Mhz DDR3 if possible, or at least 1600Mhz.

I think the concept of buying more expensive memory for a cheap, low-cost build to get better graphics performance is a little silly. But spending an extra $35 isn't too bad.

WhyteRyce fucked around with this message at 20:02 on Jun 30, 2011

# ¿ Jun 30, 2011 20:00

WhyteRyce: Dec 30, 2001

AMD fired their Senior Vice President and General Manager Rick Bergman, yikes
http://www.marketwatch.com/story/amd-says-products-gm-bergman-is-leaving-company-2011-09-22

Charlie Demerjian is in full disaster mode
http://semiaccurate.com/2011/09/23/analysis-rick-bergman-leaving-amd-has-no-up-side/

# ¿ Sep 23, 2011 19:46

WhyteRyce: Dec 30, 2001

If ARM ever gets a significant foothold in the mainstream desktop or mobile PC market, I wonder if that will change the whole regulatory/monopoly argument for AMD

# ¿ Oct 1, 2011 03:13

WhyteRyce: Dec 30, 2001

trandorian posted:

Yeah, the fact they phrased it that way made me worry too.

The way they talked up the value proposition against Intel, too.

I do like the "Please add us to your rotation guys! Please!!"

movax posted:

Desperate attempt to draw traffic to a site?
Have they said what SB this mates with yet?

Best comment from the link

quote:

Cool! Very excited for the NDA lift. Sadly, I put my computer fund into AMD stock, so I'm not sure if I'm going to be able to afford a new computer

...

:-/

Computer nerds playing at being daytraders :allears:

WhyteRyce fucked around with this message at 17:02 on Oct 6, 2011

# ¿ Oct 6, 2011 16:58

WhyteRyce: Dec 30, 2001

Really surprising considering the HardOCP guy alluded to buyers being happy with BD if they overclock.

If these numbers are close to being remotely true then that JFAMD guy will never be able to show his face on boards again after all his "these aren't official benches and should be regarded as fake!" talk the past two months in response to "leaks"

# ¿ Oct 9, 2011 06:06

WhyteRyce: Dec 30, 2001

A possible explanation for some of the lackluster leaked performance numbers

http://www.xtremesystems.org/forums/showthread.php?275786-AMD-FX-8150-Bulldozer-finally-tested&p=4969164&viewfull=1#post4969164

quote:

Actually, we already have such an issue known for Bulldozer, and NO bench-marked system has the patch installed!

The shared L1 cache is causing cross invalidations across threads so that the prefetch data is incorrect in too many cases and data must be fetched again. The fix is a "simple" memory alignment and (possible)tagging system in the kernel of Windows/Linux.

I reviewed the code for the Linux patch and was astonished by just how little I know of the Linux kernel... lol! In any event, it could easily cost 10% in terms of single threaded performance, possibly more than double that in multi-threaded loads on the same module due to the increased contention and randomness of accesses.

Not sure if ordained reviewers have been given access to the MS patch, but I'd imagine (and hope) so! Last I saw, the Linux kernel patch was still being worked on by AMD (publicly) and Linus was showing some distaste for the method used to address the issue. One comment questioned the performance cost but had received no replies... but you don't go re-working kernel memory mapping for anything less than 5-10%... just not worth it!

quote:

This patch provides performance tuning for the "Bulldozer" CPU. With its
shared instruction cache there is a chance of generating an excessive
number of cache cross-invalidates when running specific workloads on the
cores of a compute module.

This excessive amount of cross-invalidations can be observed if cache
lines backed by shared physical memory alias in bits [14:12] of their
virtual addresses, as those bits are used for the index generation.

This patch addresses the issue by zeroing out the slice [14:12] of
the file mapping's virtual address at generation time, thus forcing
those bits the same for all mappings of a single shared library across
processes and, in doing so, avoids instruction cache aliases.

It also adds the kernel command line option
"unalias_va_addr=(32|64|off)" with which virtual address unaliasing
can be enabled for 32-bit or 64-bit x86 individually, or be completely
disabled.

This change leaves virtual region address allocation on other families
and/or vendors unaffected.

# ¿ Oct 12, 2011 01:37

WhyteRyce: Dec 30, 2001

BlackMK4 posted:

I can barely believe they released this thing.

At some point with a project that has been delayed so long you have to decide to either scrap it or just release it as is and hope to recoup whatever you can.

WhyteRyce fucked around with this message at 06:44 on Oct 12, 2011

# ¿ Oct 12, 2011 06:41

WhyteRyce: Dec 30, 2001

So after all this, I wonder what the hell was the point of JFAMD posting all over the place. All he seemed to do was either condescendingly trash and discredit leaked benchmarks as being completely false when they turned out to be pretty close to the mark and post deliberately vague information and statements which ended up being completely wrong (no delay for BD, IPC improvement over and over again).

I doubt if he's ever heard again on any of those enthusiast boards because he seems to have drawn the ire of just about everyone.

# ¿ Oct 12, 2011 06:51

WhyteRyce: Dec 30, 2001

haha, apparently some poster on anandtech actually asked the AMD guy all the way back in January about potential OS issues with handling modules vs. cores

quote:

The OS doesn't know about modules, it only sees cores. But all cores are physical cores, so it won't matter.

quote:

So, here's the deal - we are obviously working with the OS and app vendors. My comment was pointed at the people who are obsessing about "how do I over-ride what you are doing because I think I am smarter and I know how to do it better."

Every environment is different, but there are people that believe they will get some massive boost by threading over modules vs. just loading threads in order. The reality is that the OS is going to figure out the best place to put the next thread. when you start up an app you may have all of the cores fire up, but once that happens, they all free up for different times. So people will never see that perfect world, and more importantly, the performance delta, for most apps, is not going to be radically different.

"Trust us we got this poo poo figured out"
...10 months...
"Here is an emergency kernel patch please use this"

# ¿ Oct 12, 2011 15:09

WhyteRyce: Dec 30, 2001

Devian666 posted:

The site actually says 40% under specific conditions and not across the board.

In relation to the core usage in windows 7.

This is about spreading FPU or SSE around between the cores rather than clogging the APU on one core while another is idle. If the patch works there will be some improvement but it'll be completely dependent on the application/game.

Why is Windows 7 getting poo poo when Linux needed a kernel patch as well?

# ¿ Oct 16, 2011 05:54

WhyteRyce: Dec 30, 2001

Ex-AMD engineer tries to explain (partially) what happened with Bulldozer
http://www.xbitlabs.com/news/cpu/display/20111013232215_Ex_AMD_Engineer_Explains_Bulldozer_Fiasco.html

quote:

Cliff A. Maier, an AMD engineer who left the company several years ago, the chip designer decided to abandon practice of hand-crafting various performance-critical parts of its chips and rely completely on automatic tools. While usage of tools that automatically implement certain technologies into silicon speeds up the design process, they cannot ensure maximum performance and efficiency.

"The management decided there should be such cross-engineering [between AMD and ATI teams within the company] ,which meant we had to stop hand-crafting our CPU designs and switch to an SoC design style. This results in giving up a lot of performance, chip area, and efficiency. The reason DEC Alphas were always much faster than anything else is they designed each transistor by hand. Intel and AMD had always done so at least for the critical parts of the chip. That changed before I left - they started to rely on synthesis tools, automatic place and route tools, etc.," said Mr. Maier in a forum post noticed by Insideris.com web-site.

Apparently, automatically-generated designs are 20% bigger and 20% slower than hand-crafted designs, which results in increased transistor count, die space, cost and power efficiency.

"I had been in charge of our design flow in the years before I left, and I had tested these tools by asking the companies who sold them to design blocks (adders, multipliers, etc.) using their tools. I let them take as long as they wanted. They always came back to me with designs that were 20% bigger, and 20% slower than our hand-crafted designs, and which suffered from electro-migration and other problems," the former AMD engineer said.

This was actually posted a long time ago but xbit is digging it back up because it is relevant now in hindsight I guess. He hasn't worked for AMD in some time and actually sounds like a pretty bitter, disgruntled dude.

Some people in the comments were joking about how this guy is just mad that he was replaced by a robot :v:

WhyteRyce fucked around with this message at 06:08 on Oct 17, 2011

# ¿ Oct 17, 2011 05:50

WhyteRyce: Dec 30, 2001

It sounded like a mix of exaggeration and disgruntled employee. If you follow around his posts you'll see him (repeatedly) make the claim that all the engineers from the golden days are no longer with the company, which is probably more telling of BD issues. Although that too sounds like more disgruntled, ex-employee complaining.

# ¿ Oct 17, 2011 17:01

WhyteRyce: Dec 30, 2001

Just in case you were still interested in Bulldozer
http://scalibq.wordpress.com/2011/10/19/amd-bulldozer-can-it-get-even-worse/

quote:

A number of reviewers have reported problems with a Blue Screen Of Death on AMD�s Bulldozer, even with stock settings:

...

As you can see, Total War: Shogun 2 triggers this bug. Which is rather ironic, since it is part of AMD�s Gaming Evolved program.

The hardware.fr review says that AMD has been able to reproduce the bug, and is working on a fix.

Now, what is happening here? Let�s look at Microsoft�s site for an explanation of this particular error code:

The CLOCK_WATCHDOG_TIMEOUT bug check has a value of 0�00000101. This indicates that an expected clock interrupt on a secondary processor, in a multi-processor system, was not received within the allocated interval.

..
Cause

The specified processor is not processing interrupts. Typically, this occurs when the processor is nonresponsive or is deadlocked.

# ¿ Oct 19, 2011 19:01

WhyteRyce: Dec 30, 2001

Obama visited the Intel Oregon facilities and Intel's CEO is on his jobs council. I don't think the administration views Intel as a big evil monopoly that needs to be broken up sans AMD

# ¿ Oct 21, 2011 00:14

WhyteRyce: Dec 30, 2001

Splitting the chipset business off from CPU makes no sense but that might just be my bias speaking.

# ¿ Oct 21, 2011 04:11

WhyteRyce: Dec 30, 2001

The Linux kernel patch has been benchmarked
http://openbenchmarking.org/result/1110200-AR-BULLDOZER41

:toot:

# ¿ Oct 21, 2011 15:41

WhyteRyce: Dec 30, 2001

cinder posted:

Has AMD directly addressed the questionable performance of BD versus their existing offerings? I had read through the previously linked thread mentioning AMD's desire to answer questions directly from the enthusiast crowd and I'm interested to see their responses, but it doesn't seem like it has been posted yet.

There's a thread on the HardOCP forums where Kyle is asking for questions that will be answered by AMD as part of an interview article.

# ¿ Oct 24, 2011 15:05

WhyteRyce: Dec 30, 2001

quote:

AMD might make chips based on both the x86 and ARM designs, some experts have speculated. But if it switches exclusively to ARM, it would leave Intel essentially alone in the x86 business, which "would make Intel kind of look like they are the guy that missed the meeting," said tech analyst Rob Enderle.

I love tech analysts

# ¿ Nov 29, 2011 16:29

WhyteRyce: Dec 30, 2001

Install Gentoo posted:

Well desktops have been outsold by notebooks since 2005 or so and if I remember right they're currently about 65% of computer sales, might even be 70% or more.

And frankly AMD has done even worse in laptops than they have in desktops, since the whole power usage problem is especially apparent and annoying when you have a battery and that lasts shorter and thing you're touching directly that heats up more.

Intel is going to be pushing ultrabooks hard, too

# ¿ Nov 30, 2011 04:15

WhyteRyce: Dec 30, 2001

What did AMD need all those marketing folks for? I only ever saw adds for them on hardware sites and as part of game intros. And those banners looked like something a 14 year old with a pirated copy of PS could come up with

# ¿ Dec 3, 2011 23:17

WhyteRyce: Dec 30, 2001

I bought one for my HTPC and I wish I just stuck with my old C2D/Ion setup

# ¿ Dec 6, 2011 01:09

WhyteRyce: Dec 30, 2001

Oops, sorry I should say I have a E-350. It's not where near as rock solid as the other setup was. I've got some weird HDMI issue where occasionally the resolution looks like it gets set really low (i.e. my WMC looks like it's running 1024x768 stretched to fit my screen), which goes away if I minimize and re-maximize. Which is super-annoying in a HTPC setup that you want to control only with a remote. If I elect to connect directly to my TV instead of through my receiver, then my screen will turn black randomly while idle. WMC will also crash if I have too many files in a video directory. It also crashes when it tries to render the thumbnail for certain MKVs. Continually waiting around for Silverlight 5 so I can do HD Netflix is also fun. And AMD removed the overscan correction tool in a couple of their driver releases.

WhyteRyce fucked around with this message at 01:37 on Dec 6, 2011

# ¿ Dec 6, 2011 01:34

WhyteRyce: Dec 30, 2001

movax posted:

Not to derail too much, but this is why I left HTPCs behind. I know a lot of people have them running successfully and love them to death, but I just had so much trouble with them I went back to dedicated set-top media boxes. Sacrifice in broader software compatibility, but much less painful. It didn't help that my target display was connected via 1080i component, which means a fun battle against overscan.

That said, maybe some AMD hardware will be finding its way into aforementioned boxes, but they've got stiff competition from the existing players in the field.

I really contemplated switching to a PS3 or something like a Boxee, but I planned on getting a network CableCard tuner and that thing was too enticing.

Don't get me wrong, I really like my current HTPC. Having an all-in-one device for my movies, Cable TV, and Blu-Ray is loving awesome. When my HTPC is working, it's loving amazing. It's just that, when it doesn't work it's extremely aggravating. I'm assuming what I'm hitting is just stupid AMD driver/software related issues, that didn't effect my old ION platform.

The real annoying thing with this is that AMD and MS just tend to point figures at each other in situations like this and you won't ever get a real resolution. The RTC time drift is really annoying too. I've never had a platform which has this much drift in this short amount of time, which really screws with my DVR programming if I let it get out of hand. I setup a Windows task to sync it every hour, but I guess it's not working since I still drift off a over a minute within a week.

I guess that's why I'm so annoyed at my e-350 board. So many small annoying issues

WhyteRyce fucked around with this message at 07:04 on Dec 6, 2011

# ¿ Dec 6, 2011 06:57

WhyteRyce: Dec 30, 2001

The thing that really annoyed me with the e-350 when I was shopping around, most of the boards came with some lovely, tiny, noisy fan on it. Very little in the way of true silent cooling options, except for an ASUS board that had a giant passive heatsink on it. But a bunch of reviews on Newegg claimed it still overheated.

# ¿ Dec 6, 2011 07:17

WhyteRyce: Dec 30, 2001

Oh the days of picking out a Matrox 2D card for the ultimate pairing with a Voodoo.

# ¿ Jan 22, 2012 20:46

WhyteRyce: Dec 30, 2001

I think every video card maker on the planet had their own optimized version of Mechwarrior 2

# ¿ Jan 22, 2012 21:01

WhyteRyce: Dec 30, 2001

Nostrum posted:

And in the cruelest tradition of irony, PowerVR's SGX mobile chips absolutely crush the relatively disappointing GeForceULP portion of Tegra 2/3.

PowerVR saw the light 15 years ago. Everything after that up until now was them just playing possum :ninja:

WhyteRyce fucked around with this message at 01:30 on Jan 24, 2012

# ¿ Jan 24, 2012 01:22

WhyteRyce: Dec 30, 2001

movax posted:

Man, the comments section for the PCIe 3.0 article were pretty inducing.

PCIe is a terrible interconnect and we should all be using Hyper Transport instead but it doesn't matter because Intel controls the PCI SIG so we are stuck with this :words:

# ¿ Jun 4, 2012 05:42

WhyteRyce: Dec 30, 2001

keyvin posted:

AMD Is preparing to lay off 20-30% of their workforce: http://allthingsd.com/20121012/exclusive-amd-to-cut-up-to-30-percent-of-workforce/.

I like the guy in the comments who was being totally serious about trying to get a hold of ARM management to give them advise

# ¿ Oct 14, 2012 03:50

WhyteRyce: Dec 30, 2001

Install Gentoo posted:

Do you have any idea how much money it took to get Intel or AMD processors to the performance they are today? $100 billion is nothing in comparison, and frankly the A6 doesn't show poo poo as far as getting an ARM based architecture to x86-64 performance in laptop/desktop applications.

Also just throwing a pile of money at something is not guaranteed to get results. Although that is a lot of money.

I figured if Apple were to throw that money at something, it would be at fabs.

WhyteRyce fucked around with this message at 04:56 on Oct 15, 2012

# ¿ Oct 15, 2012 04:53

WhyteRyce: Dec 30, 2001

Professor Science posted:

the flat cost of x86 instruction decode is pretty minimal these days

You mean those guys who post in the comments on Engadet and Anandtech are wrong??

# ¿ Oct 15, 2012 16:33

WhyteRyce: Dec 30, 2001

Alereon posted:

By "busted" I guess you meant "confirmed", as Medfield solidly proved that it will take a completely new microarchitecture for Intel x86 offerings to be competitive with ARM.

No one was talking about performance of Medfield vs. ARM's latest offerings, just that opinion of a lot of arm chair architects seemed (or still seems) to be that x86 is just too inefficient and wasteful to be ever used in a small, low power applications.

# ¿ Oct 16, 2012 16:21

WhyteRyce: Dec 30, 2001

Star War Sex Parrot posted:

AMD's not going any lower right? I'm safe to go long on them now?

I remember when their stock first fell to $7, tons of pretend stock market gurus on hardware boards were telling everyone to buy it up because that was a bargain price and had no where to go but up :allears:

# ¿ Nov 9, 2012 06:36

WhyteRyce: Dec 30, 2001

http://allthingsd.com/20121113/amd-exploring-options-including-breakup-sale/

quote:

Shares of the chipmaker Advanced Micro Devices are up by more than 14 percent on a report by Reuters saying that the company has hired J.P. Morgan to �explore options� that could include a disposition of its patents or an outright sale.

Well things seem to keep getting better for AMD. I'm assuming they are looking to sell off some of their assets. Wonder how much longer they can keep going at this rate

WhyteRyce fucked around with this message at 05:04 on Nov 15, 2012

# ¿ Nov 15, 2012 04:50

WhyteRyce: Dec 30, 2001

icantfindaname posted:

Considering Intel is an American corporation, then America?

Intel is a huge multi-national company

# ¿ Nov 29, 2012 20:02

WhyteRyce: Dec 30, 2001

icantfindaname posted:

So I guess that means it's above the law of any one country, because it has operations in lots of them? I understand the point, I'm just saying that it's not as big a problem as it's made out to be, besides the fact that it will never happen, I suppose.

And again, I don't think nationalizing Intel is a good idea. I think public funding for cpu research is an interesting idea, but I'm not knowledgeable enough to say for sure. I think I misunderstood your position against nationalizing Intel in the first place.

I'm not talking about the law or any of that monopoly stuff people are throwing around, just that you're talking about nationalizing a company that has a good number of its people and resources in other countries.

# ¿ Nov 29, 2012 20:24

WhyteRyce: Dec 30, 2001

Even at the peak of their popularity, consumers are dumb and didn't know any better. Only nerds gave a poo poo about benchmarks and a sizable portion of buyers probably stuck to what brands they recognized. It didn't help that earlier non-Intel chips were poo poo. I had a hell of a time convincing people I knew who trusted my opinion to buy AMD at the time, I can't even imagine what it was like for your average consumer asking the Circuit City guy for help.

# ¿ Aug 19, 2014 17:00

WhyteRyce: Dec 30, 2001

cisco privilege posted:

What alot of people like to forget is that AMD chipsets from the era were loving awful with missing or misfiring features compared to Intel chipsets. Sure, a Pentium 4 wouldn't run nearly as well, but when AMD boards had VIA or SiS-level quality concerns and weird compatibility issues it usually wasn't worth trying them out compared to an Intel setup which would presumably just work, albeit with measurably worse performance.

Apple switched to Intel during AMD's hayday as well. This is a totally blind guess to me but I assume things like that and superior supply chain management are more important to these guys than just CPU benchmarks

# ¿ Aug 20, 2014 17:29

Adbot: ADBOT LOVES YOU

# ¿ May 4, 2024 06:21

WhyteRyce: Dec 30, 2001

Don Lapre posted:

The big deal with intel was performance per watt.

OEMs probably don't care about a regular consumer's power bill and your average Joe consumer had no concept of what that was 10 years ago

# ¿ Aug 20, 2014 17:35

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > AMD CPU and Platform Discussion: Ryzen Might Be...Good? *knocks on wood*

«‹›3 »