Weekend Shortbread

Eight is Enough

  1. Neowin: Windows 8.1 ‘August update’ BSODs affected less than 0.01% of users
  2. techPowerUp!: Intel Core i7-5820K features fewer PCI-Express lanes after all
  3. LinuxGizmos: Android-on-ARM mini-PC draws less-than-7W
  4. CHW.com: AMD working on 20nm “Faraway Islands” GPU (in Spanish)
  5. WCCFtech: AMD Catalyst 13.9 WHQL to 14.7 RC benchmark

    comparison – 4% overall increase with interesting anomaly’

  6. Reuters’ exclusive: Apple iPhone 6 screen snag leaves supply chain scrambling
  7. Mac Rumors: iPhone 6 information leak sourced from

    Foxconn details dimensions of 4.7″ and 5.5″ versions

  8. GameSpot: Witcher 3 dev responds to report that

    game may not run in 1080p on PS4, Xbox One


Weekend

  1. CPU World: Intel “Braswell” systems-on-a-chip delayed (again)
  2. Celebrating 30 years of graphics and gaming at AMD
  3. Newegg’s weekend deals
  4. Dealzon’s deals: $50 coupon 17.3″ 1080p touch Dell 5000 series

    i7-4510U, $50 coupon for 14″ Lenovo Y410p i7-4700MQ / GT

    755M, and $100 off 13.3″ 1080p touch Dell XPS 13

Systems, storage, and networking

  1. ThinkComputers has a hands-on with the ECS LIVA mini PC kit
  2. Hardware Slave’s home server build part 3: the context
  3. Ocaholic’s Asus Maximus VII Impact review
  4. TweakTown’s Asus P9A-I/C2550/SAS/4L (Intel

    Avoton) server motherboard reviewโ€

  5. Hardware Heaven’s 960GB Kingston V310 SSD review
  6. CowcotLand reviews 256GB Plextor M6 Pro SSD (in French)
  7. APH Networks on QNAP HS-210 NAS
  8. Hardware Heaven’s Netgear EX6200 – AC1200 extender review

Multimedia and cooling

  1. Benchmark Reviews on Roccat Kave XTD 5.1 headset
  2. KitGuru’s i-Rocks K10 USB keyboard review
  3. techPowerUp! reviews Zalman Reserator 3 Max Dual
Comments closed
    • Kurkotain
    • 5 years ago

    So let me get this straight.

    Semi Accurate is whining about a -lack of accuracy- and fact checking in tech journalism?

    A moment to gather my thoughts…

    HAHAHAHA

    • ronch
    • 5 years ago

    Just thought I’d share this.

    [url=http://www.tomshardware.com/reviews/nvidia-tegra-qualcomm-snapdragon-apple-samsung-exynos,3910.html<]An Introduction To The Major ARM-Based SoCs[/url<]

    • ronch
    • 5 years ago

    [quote<]AMD working on 20nm "Faraway Islands" GPU (in Spanish)[/quote<] Does anyone expect AMD to shrink their FX chips to 20nm or similar anytime soon, if at all?

    • UnfriendlyFire
    • 5 years ago

    “WCCFtech: AMD Catalyst 13.9 WHQL to 14.7 RC benchmark comparison – 4% overall increase with interesting anomaly”

    Quick question, is the 14.7 RC compatible with mobile switchable graphics. I just bought a laptop with a Radeon 8750M, and it has a graphic driver in fall 2013.

    • jdaven
    • 5 years ago

    Scott was just shown on the live streaming event from AMD just now talking about his favorite Radeon product the 9700 Pro.

    Chuckula, does that make TR AMD bias just like Anandtech?

    • jdaven
    • 5 years ago

    CPU World: Intel “Braswell” systems-on-a-chip delayed (again)

    Let the record show that both 14 nm from Intel and 20 nm from TSMC are equally delayed and will have the same three year time interval between mass production and market availability. Intel will be ahead by one process node just like it’s always been for the last ten years. Intel is neither gaining new ground or falling further behind.

      • NeelyCam
      • 5 years ago

      Braswell is not Broadwell

        • jdaven
        • 5 years ago

        I know. It’s just another 14 nm processor family that is getting delayed.

          • NeelyCam
          • 5 years ago

          Yes, but if Broadwell is coming out at the end of this year, your argument about Intel 14nm doesn’t hold water.

          Also:

          [quote<]Let the record show that both 14 nm from Intel and 20 nm from TSMC are equally delayed and will have the same [b<]three year time interval between mass production and market availability[/b<].[/quote<] What do you mean with this "three year time interval between mass production and market availability"? It certainly doesn't take three years to get products on the market after mass production has started.. Or do you mean three years between market availability of Intel 22nm->14nm and TSMC 28nm->20nm? Because to me it looks like Intel's transition takes 2.5 years, while TSMC's takes 3.5 years - i.e., the gap is increasing.

            • jdaven
            • 5 years ago

            I was trying to combine mass production and market availability as a single event. There are only months of separation between the two and wasn’t my meaning or point. Sorry for the confusion. I should have just said market availability because that is where we have definitive dates. And here they are:

            Jan 31, 2012 – Release of Radeon HD 7950 at 28 nm. First mainstream consumer product that I could find on 28 nm
            Sept 2014 to Mar 2015 – Timeframe for the release of 20 nm products for various products including SoCs and GPUs. Let’s just say Jan 2015.

            So that’s three years. Now for Intel.

            April 2012 – Release of desktop and laptop Ivy Bridge processors
            Jan to June 2015 and beyond – Estimated release of desktop and laptop Broadwell processors. Let’s just say Mar 2015.

            So that’s three years. But what about Core M. There hasn’t really been a product like this before which is why it has a new name. So it’s three years between Core iX products.

            Now we can debate dates all day. I can say 20 nm SoCs from TSMC in the new iPhone next month is the start date. You can say Core M in the October to December time frame is the start date of 14 nm. Either way its still the same time interval between the two process transitions. It’s definitely going to be three years between 28 to 20 nm GPUs and 22 to 14 nm Core iX processors.

            • NeelyCam
            • 5 years ago

            [quote<]Jan to June 2015 and beyond - Estimated release of desktop and laptop Broadwell processors. Let's just say Mar 2015.[/quote<] Core M products are expected to come out by the end of the year. Your March 2015 and the excuse that Core M doesn't count is BS. I didn't realize 28nm products didn't come out until 2012, though, so 28nm->20nm is around 2.5-3 years then. I guess TSMC has been going with 2.5-3 year release cycle longer than I thought - 40nm came out around mid 2009.. But recall how Intel was running at a 2-year cycle consistently for a long time? The slowdown happened when FinFET was added. What do you think will happen to TSMC when they try to add FinFETs...?

            • jdaven
            • 5 years ago

            I said you can count 14 nm Core M if you also count the September launch of the iPhone that will be using TSMC’s 20 nm process. Since both 22 nm and 28 nm came out at the exact same time, that means TSMC got to the 20 nm process before Intel got to the 14 nm process. The time interval slightly favors TSMC.

            Why is that so hard to understand?

            Edit: So I guess you can say that TSMC is right on schedule with 2.5-3 year releases and Intel is falling behind with a 2.5-3 year release.

            • Ninjitsu
            • 5 years ago

            [quote<] if you also count the September launch of the iPhone that will be using TSMC's 20 nm process. [/quote<] That's not confirmed, though. If TSMC is seeing significant demand in Q4, I don't think 20nm can come out in huge volume in Q3. But it's [i<]expected[/i<] to be 20nm, though I've not read anything beyond speculation so far. Core M, on the other hand, has been officially announced as a 14nm part.

            • NeelyCam
            • 5 years ago

            [quote<]Since both 22 nm and 28 nm came out at the exact same time[/quote<] Your earlier comment stated that Intel 22nm came out in April 2012, while TSMC 28nm came out in January 2012. If you insist on continuing this silly argument, at least be consistent about it.

            • jdaven
            • 5 years ago

            Really? You are calling me out on the difference between Jan 31, 2012 and April 2012. Two months out of two to three years. Is this how you are going to attack my arguments? Nick picking every little detail?

            Jeez, I should have just said first half of 2012. Is this okay your lordship?

            • NeelyCam
            • 5 years ago

            [quote<]Is this okay your lordship?[/quote<] No. You shall call me "my Lord" or "your Highness". If this disrespect does not end, my knights shall separate your blasphemous head from the rest of your blasphemous body.

            • UnfriendlyFire
            • 5 years ago

            The higher end Broadwell chips are being delayed.

            Apple mentioned that they won’t be releasing Broadwell Macbooks until around spring 2015, and they have every reason to release a Broadwell laptop ASAP.

            There were also rumors that Intel has delayed the Broadwell desktop chips to fall 2015.

            • Ninjitsu
            • 5 years ago

            [quote<] There were also rumors that Intel has delayed the Broadwell desktop chips to fall 2015. [/quote<] Wasn't that Braswell, though? The rest of the Broadwell line up seems to be launching Jan onwards. Apple's own internal validation etc. will take time, i guess.

            • jdaven
            • 5 years ago

            Here is the latest on Broadwell desktop chips:

            [url<]http://www.xbitlabs.com/news/cpu/display/20140714225810_Intel_Postpones_Launch_of_High_End_Broadwell_K_Processors_to_July_September_2015.html[/url<] It's not a conspiracy theory but leaked production roadmaps. Give whatever weight to them you want.

            • MadManOriginal
            • 5 years ago

            I would be surprised if TSMC is able to produce 20nm chips in the volume needed for an Apple product launch in September. Supposedly there is a supply issue with the iPhone display, so separating that from any 20nm chip supply issue and the ‘omg Apple sells out due to demand’ will be tough unless someone leaks specifics. The fact that 20nm GPU launches are only far-off rumors and not imminent ‘next month’ rumors doesn’t bode well for TSMC 20nm having volume for a September iPhone launch.

            • jdaven
            • 5 years ago

            Then this report should surprise the heck out of you:

            [url<]http://9to5mac.com/2014/08/25/tsmc-expected-to-begin-16nm-volume-production-for-apples-a9-chip-in-early-2015/[/url<] Money can do wonders so it seems.

            • blastdoor
            • 5 years ago

            If 16 nm or 14 nm chips from TSMC or samsung end up in iPhones next year there are going to be heads exploding all over TR forums. Not to mention massive down voting for every jdaven post on any subject until the end of time.

            • jdaven
            • 5 years ago

            Calling out changing market trends in a room full of fanboys can be a dangerous endeavor. But damnit someone’s got to do it. ๐Ÿ™‚

            • blastdoor
            • 5 years ago

            I’m tempted to down vote you, but there’s no way to indicate I mean it as an ironic down vote so I guess I won’t do it ๐Ÿ™‚

      • MadManOriginal
      • 5 years ago

      I would love to see what happens if Intel decides to fab iPhone chips under contract on their most advanced node. Since jdaven has selectively chosen to only care about iStuff and not Macs which use Intel chips, would his head explode if iStuff chips were made by Intel as well?

        • jdaven
        • 5 years ago

        And I would love a million dollars. Neither of which is ever going to happen.

        • blastdoor
        • 5 years ago

        Ha ha ha ha — “if Intel decides” — ha.

    • ronch
    • 5 years ago

    [quote<]AMD working on 20nm "Faraway Islands" GPU (in Spanish)[/quote<] Well, I hope Faraway Islands' release date wouldn't be so far away.

      • The Dark One
      • 5 years ago

      Their source is Charlie Demerjian, so take it with [s<]a[/s<] [i<]several[/i<] grains of salt.

        • Deanjo
        • 5 years ago

        So the code name is “Fantasy Island” then right?

        • Alexko
        • 5 years ago

        No, it isn’t.

          • NeelyCam
          • 5 years ago

          Interestingly, it sort of was, but information was faked

    • Meadows
    • 5 years ago

    About the Witcher 3 resolution nonsense: I appreciate the crowd control regarding an employee’s “careless” statement. On a different note however, it’s kind of sad how the new consoles are barely a year old yet developers have struggled with their apparent limitations in many titles already.

    I guess they thought these consoles would provide the performance of today’s average desktop gaming PC and they were far, far off the mark. Shame on you, AMD.

    Also, I don’t keep up with console news very well, but can someone explain why the XBox is slower?

      • sweatshopking
      • 5 years ago

      It has slower ram, and more complicated to use. It also has a slower video card. It has a faster CPU however, and the potential for some interesting things in the future with cloud based rendering, but its not there yet.

      • Prestige Worldwide
      • 5 years ago

      Shame on AMD? More like shame on MS and and Sony for cheaping out. AMD gave them what they asked for.

        • Pwnstar
        • 5 years ago

        Exactly. It’s not AMD’s fault that those companies went with a minimum of performance. They could have doubled the Ghz of the CPU and went with a discrete nVidia 760m GPU but they didn’t.

        • willyolio
        • 5 years ago

        no no, it’s totally AMD’s fault. It’s also AMD”s fault that they are delivering slower chips to microsoft and giving better ones to Sony. it’s not like they had them designed that way or anything.

          • UnfriendlyFire
          • 5 years ago

          MS and Sony could’ve went to Intel and/or Nividia instead of AMD.

          And if AMD really under-delivered, we would’ve heard something from MS and/or Sony by now about “AMD failing to uphold their part of the contract”.

          There are some PSU OEM manufacturers that build really high quality stuff or Leadman/Powmax-quality stuff. Why? Because it came down to the brand purchaser of what kind of PSUs they wanted to buy and slap their labels on.

            • Godel
            • 5 years ago

            NVidia have previously stated that they didn’t want to go into the OEM console market for Sony and MS, and Intel products have relatively crap video performance.

            The XBox One was partially slowed by the initial compulsory inclusion of the Kinect, which I understand used up a bunch of CPU cycles.

            • Pwnstar
            • 5 years ago

            That was sour grapes from nVidia after they lost both contracts to AMD. You can bet that if they had won either of the contracts instead, they would not have said that.

            • willyolio
            • 5 years ago

            wow and i thought i couldn’t make the sarcasm more obvious.

Pin It on Pinterest

Share This