355. John Peddie (NVIDIA) declaration
- At all times, please reference the page # and line # (line numbers on the left side of the document) on the PAGE itself (not PDF page count)
- Please BE SPECIFIC and FACTUAL when stating your comments.
- If you are referencing another document, please state the file name, page number, line numbers, etc.
- Master PDF file is here (nvidia.355.jon peddie declaration)
{ 11 comments… read them below or add one }
Milberg has asserted that depreciation is not a factor in the settlement, yet Peddie brings up the fact that our computers are depreciating. Not to mention, Moore’s law was the number of transistors on a die doubles every two years…not performance. Page 24 of 31 (exhibit C) of the Peddie declaration, he even says that the AMD V140 is a worse CPU than the class computers. And that’s after 4 years of “moore’s law”. ***PAGE 25 of 31 of Peddie’s Declaration…HE HAS BENCHMARKS SHOWING THAT THE REPLACEMENT CPU IS AT TIMES ~65% OF THE OLD CPUS
Also, ‘reliability’ is of no consequence to us. If we take care of our hardware, it should last long enough…
Peddie also mentions that after a “quick search” he has come up with a $370 dv6000. Okay, well after my quick search, here’s a dv6308nr for $699.99 on Amazon: Amazon.com: HP Pavilion dv6308nr – Turion 64 X2 TL-50 / 1.6 GHz – RAM 1 GB – HDD 120 GB – DVD?RW (+R DL) / DVD-RAM – GF Go 6150 – WLAN : 802.11b/g – Vista Home Premium – 15.4″ Widescreen TFT 1280 x 800 ( WXGA ) – remarketed: Computer & Accessories
http://www.amazon.com/HP-Pavilion-dv6308nr-Widescreen-remarketed/dp/B000TZT96M is the link as of 3/5/2010.
bottom of page 8, top of page 9: He states Mr. Vlastone is ignorant of Moore’s law, which he states as “technological performance increases by a factor of 2 every 2 years. In other words, a person who buys a computer can buy a new computer with the same performance two years later for half the price”
NOT TRUE – he is showing his own ignorance here. Moore’s law is “The quantity of transistors that can be placed inexpensively on an integrated circuit has doubled approximately every two years”
– Gordon E. Moore, “Cramming more components onto integrated circuits”, Original paper obtained via link at Wikipedia (footer note 1))
And obviously, if the cost part was true, we’d be paying about a dollar for a notebook by now.
“Largely forgotten is that Moore, in an attempt to keep his ‘law’ effective, formulated
Moore’s Second Law, which says that facility costs increase on a semi-log scale. In other
words, every additional doubling of device complexity increases manufacturing costs
exponentially.”
– Jeff Dorsch “Does Moore’s Law Still Hold Up?” another Wikipedia footnote document. Can also find many other references to the fact that Moore’s law causes prices to RISE for smaller components
Uses ‘or’ instead of the correct ‘and’ between kind and equal. pg2 L19; pg3L10;
Seems from pg2 L19 that the Asus tablet netbook was selected by NVIDIA and not Mr. Peddie.
Stresses ‘holistic’ approach to system evaluation. One of the best ‘holistic’ evaluations of netbooks was published by Intel, the maker of computer systems and subsystems. http://www.intel.com/consumer/products/netbook.htm
pg3 L23 makes an assumption that tablets only run ‘internet’, office products, etc. The TX1000 series is an Entertainment convertible tablet/laptop. It has a larger higher resolution screen and a dual core processor and higher memory capacity to run more complex and cpu intensive programs and multi-tasking with multi-monitors.
pg4 L1,2,3,4 list ‘advantages’ over the TX1000 series. These are reasons someone would by one of the tiny Sony or Fujitsu ultra portables that were available at the time I purchased my TX1410us. I bought a the larger more powerful TX1000 series for the multi-media power, larger screen, optical drive (DVD), and multiple ports, even though it had less battery life.
pg4 L6 There is no backing to the statement “It also has a more efficient and readable display”. TX1000 has a 21.1″ 1280×800 display while the netbook has a 10.1″ 1024×768 display.
pg4 L7 The version of the newer Windows operating system is a crippled version. Not capable of multi-tasking, dual display, and many other features that are present in my ‘older’ Windows operating system. We should not be penalized because council took so long is determining this settlement.
pg4 L11,12 I was only able to get approx a year of full use out of my tablet laptop. It stopped working correctly just after the warranty period. HP would not help. I was able to get it to limp along at lower performance and greater heat for a little while longer.
pg5 L15,16,17 CQ56 is not a tablet and I will not replace my tablet with a regular laptop with a significant decrease in computing power.
pg7 L9 Our computers are not “OLD”. I only used my for approx 1 year. The council has been arguing the settlement for several years and we should not be harmed or penalized for this.
pg7 L15 Our computers performance for exchange consideration should not be penalized for not working because of the defect that this settlement is supposed to correct.
pg7 L22 The TX series is not an ultra portable convertible tablet. It is a fully functioning laptop that also has a tablet function.
pg7 L26,27 These ‘advantages’ only apply to those that want an ultra portable. The TX series is a fully functioning and powerful laptop that also has a convertible tablet feature. Much like the current HP line the TM2T which is the continuation of the same niche the TX series fit in the HP line.
pg8 L7 Again I only was able to use my system fully functional for approx 1 year.
pg8 L21 Again I only was able to use my system fully functional for approx 1 year.
pg8 L27 Again I only used my Tablet Laptop for approx 1 year and the case has been running for several years.
pg9 L2,3,4 Depreciation that was accelerated by the general view that these models have hardware problems.
Here’s a 2010 article that is pretty ironic considering Nvidia is using the Moore’s Law argument that the CQ56 must be, what, 4 times faster than my (now dead for 2 years) 4 year old dv9000 AND that the single-core CPU is better than the dual-core. It states Moore’s law no longer applies AND that multi-core architecture is the way around this “crisis”:
Nvidia’s chief scientist has claimed that Moore’s law is no longer applicable to the silicon world of CPUs.
“We have reached the limit of what is possible with one or more traditional, serial central processing units, or CPUs. CPU performance no longer doubles every 18 months. And that poses a grave threat to the many industries that rely on the historic growth in computing performance,” Bill Dally explained in a Forbes column.
“The good news is that there is a way out of this crisis. Parallel computing can resurrect Moore’s Law and provide a platform for future economic growth and commercial innovation. The challenge is for the computing industry to drop practices that have been in use for decades and adapt to this new platform.”
According to Dally, there is a “critical need” to build energy-efficient parallel computers – or throughput computers – in which numerous processing cores, each optimized for efficiency, not serial speed, work together on the solution of a problem.
“[But] more importantly, parallel computers, such as graphics processing units, or GPUs, enable continued scaling of computing performance in today’s energy-constrained environment. Every three years we can increase the number of transistors (and cores) by a factor of four.”
Note: There are two sets of page numbers on this document. I am using the blue page numbers on the top of the page. Also, the text I am quoting may be paraphrased for brevity.
pg 7, L12 “faster CPU” 5/6 of the benchmarks Bagherzadeh referenced showed that the TL-64 handily beat the V140.
pg 7, L13 “higher speed drive” I’ll assume that he meant hard disk drive and not optical drive. I thought the replacement had a 5400 RPM HDD and all of the originals had at least a 5400 RPM HDD.
pg 7, L13 “is lighter” Of course a smaller computer (15.6″ CQ56) is lighter than a larger computer (17″ dv9000). In all likelihood, people who bought a “desktop replacement” ranked battery life and weight as less important than screen size, CPU speed, and accessories.
pg 7, L25-26 “Dr. Bagherzadeh correctly notes… the CQ56 is superior in almost all major areas” As I asked in my Bagherzadeh post, what are the major areas? The CQ56 lags in screen size and resolution and processor speed, is equivalent in a couple areas, and has a better GPU and memory. I don’t know many people who buy laptops and consider the type of RAM and the GPU as more important than the screen size and processor speed.
pg 8, L3 “improved… battery life” The battery life is going to be better because the screen is significantly smaller. My laptop has an 8-cell battery, so I would expect an improved battery life from a smaller computer; however, if the CQ56 had a 17″ screen, it wouldn’t be a very significant improvement over my dv9000.
pg 8, L4 “the CQ56 is similar in kind and value to the older computers” This is the first I’ve seen an expert explicitly state “kind and value”. The CQ56-115DX is currently (as of 3/6/11) selling at $329.99. Since he stated that my 17″ dv9000 is similar in kind to the 15.6″ CQ56, I should be able to buy a 17″ laptop for about $330. However, the cheapest 17″ laptop that Best Buy sells (the HP G72-C55DX) costs $679.99. That’s 106% more than the CQ56. I also checked Walmart, which is currently selling the CQ56-219WM (which has a faster processor than the 115DX) for $298. The cheapest 17″ laptop Walmart sells costs $498 (The Acer AS7741Z-4433). That’s 67% more than the CQ56-219WM.
pg 8, L8-11 “Mr. VLastone does not take into account the fact that HP consumers are exchanging old computers for newer models” Milberg has been telling us that depreciation was not a factor in the laptop selection. Even if it was, the expert should be using the age when the laptop actually began failing, not the age of the laptop when it is being sent in to be replaced. If you use this definition, the laptops being replaced are mostly 1-2 years old, so they were not “nearing the end of their useful life” when they failed.
pg 8, L18-20 “inappropriate/impossible to compare systems feature-by-feature” This is definitely not impossible, just time consuming and expensive. Why is it inappropriate? The settlement agreement states that we will receive a computer similar to the one we own. If it’s too time-consuming to compare each laptop on a feature-by-feature basis, it shouldn’t be too hard to create a couple different groups (tablet, 14″, 15″, 17″) and the replacement in each group should be equal to or better than the best computer that the class members bought. Based on their expert’s interpretation of Moore’s law (page 10), this shouldn’t be cost prohibitive, since the original laptops are 2-5 years old, they should be able to be replaced for 50% to less than 25% of their original cost (note this is bastardization of Moore’s law).
pg 8, L25-27 “Asus has a longer battery life, has a more efficient display, is lighter weight and far more portable”. My cell phone also has those same characteristics, but that doesn’t make it suitable as a replacement.
pg 9, L4 “faster CPU, … higher speed drive, and is lighter” See my comments above as to why these are inaccurate or deceptive statements.
pg 9, L6-7 “the HP models covered by the settlement are old systems that have surpassed their expected useful lives” Based on the affected model list [http://nvidiasettlement.com/affectedmodels.html], computers purchased as recently as April 30, 2009 are part of this settlement. Have less-than-2-year-old computers really “surpassed their expected useful lives”? If you look at Exhibit B, you’ll see that 20.4% of computers had hardware malfuntions in the first 3 years of ownership. Since a hardware malfunction would be covered by warranty for at least the first year, this drops the failure rate for laptops in their second and third years to 15.7% (20.4 to 4.7). Based on the forums that I’ve been reading and my own experience, the HP laptops seem to be failing prior to even being 3 years old (with some failing in < 13 months), so there is an 84.3% chance that if we were actually able to use our laptops for 3 years, that they would be broken by a hardware malfunction (the accident rate is irrelevant because we weren't actually able to use our laptops; and laptops that were broken due to an accident likely aren't part of this settlement because they likely wouldn't have experienced the symptoms described and been able to continue using the laptop). My dv9000 has a bigger screen and a faster CPU than the CQ56, and has an 84.3% chance of not having any hardware failure (other than the defective NVIDIA GPU), but somehow it has "surpassed its expected useful life."
pg 9, L10-11 "new warranty is particularly important given the likelihood of failure in the first year of use" Based on exhibit A, there is a 4.7% chance of a hardware malfucnction in the first year. This new warranty, which less than 5% of those with a replacement laptop is likely to use, definitely does not offset the cheaper computer that 100% of the HP replacement laptop claimants will get.
pg 9, L14-19 "it is necessary to find the common attributes of a system without regard for an indvidual's preference." Then why is the the lowest-common-denominator laptop (the CQ56) selected as the replacement?
pg 9, L20-24 "the majority of systems that are being exchanged are 2-4 years old, would be at or near the end of their useful life…" Our laptops certainly weren't "near the end of their useful life" when the NVIDIA GPU failed. The majority of people I know are using laptops that are at least 3 years old, so I think it's incorrect to say that a 2-year-old laptop is "near the end of its useful life". In fact, one of the computers that I've been using since the GPU on my dv9000 started dying is an HP laptop from 2002. While it pales in comparison to my dv9000, it is sufficient for checking my e-mail, browsing the internet, or working on office documents.
pg 9, L27, "Vlastone ignores concepts of depreciation…" Jon Peddle's declaration uses the incorrect time frame when talking about depreciation. If a laptop failed after 1 year and has been sitting in a closest for 2 years, then the depreciated value should be for one year. However, the settlement agreement and class notice said nothing about depreciation.
pg 10, L1-2 "In other words, a person who buys a computer can buy a new computer with the same performance two years later for half the price." Wow, that is completely untrue. Moore's law says absolutely nothing about cost. This statement may be true if you were referring to only the processor. But certainly there are a lot of fixed costs involved with building and selling any laptop, which don't decrease because of Moore's law.
pg 10, L3-5 "The dv6258c laptop can be purchased for roughly $370, which is similar to the price of a CQ56". There are numerous problems with this statement. As I mentioned above, the CQ56 retails for $298-$330. So, I suppose you could argue that 12-24% of the purchase price is roughly equivalent. However, the dv62xx laptops have a defective NVIDIA GPU, which greatly diminishes the value of the laptop. A Google search for "dv6258c" returned no results, so I can't comment on the specifications of this computer. However, I don't believe that it comes with a 17" screen. A quick search of recently-sold dv9000 laptops on ebay showed the defective dv9000 laptops selling for $550 (67% more than the CQ56) and a used dv9000 selling for $503.51 (53% more than the CQ56). This clearly shows that the CQ56 is worth significantly less than a DEFECTIVE dv9000 that was manufactured a couple years ago (the model that sold for $550 was an RL015AV, which is part of the settlement).
what does laptop failure rate have to do with this case, anyway? We’re discussing (supposedly, since they don’t admit anything – though they really have admitted it, over the years) defective parts manufactured by Nvidia. I don’t get where the whole “well, they break anyway” argument is relevant.
I agree you with. Our apparent opposition (NVIDIA, Milberg, et. al.) brought up failure rate because the CQ56 will come with a 1 year warranty. So their argument is that a crappy laptop with a warranty (which per Exhibit B, only 4.7% of claimants will be expected to use) is equivalent to a nice laptop without a warranty.
pg 7, L12. Peddie also makes the mistake of stating that RAM in the CQ56 (DDR3-1066) is faster than the RAM supported in our laptops (DDR2-1066). See my post in the Bagherzadeh declaration for details.
pg 8, L2-3. Again references faster memory access.
pg 9, L2-3. Same mistake
Page 4, Lines 3-8. Please notice that Mr. Peddie presents no data of his own determination and has sought additional info from mostly Nvidia employees and Dr. B. You seriously have to wonder if he has even looked at either model in person. He even goes as far as using the exact $ value for replacement components that the Asus does not have. In a classroom, if one person puts “I don’t know” on a test and another writes “I don’t either”. It’s pretty dang easy to tell whose cheating….. don’t you think?
Page 5. Lines 18. If nearly a third of computers fail with the first three years doesn’t this conversely mean that over 66% run perfectly well after 3 years? Also if you have all these notebook data points, why haven’t you published the actual failure rates for affected Apple, Dell, and HP computers with these specific Nvidia chips in them? What’s not right with this picture?
Pg 5 L9-10
“The new Asus replacement model also still has a better battery life,”
http://commercial.asus.com/product/detail/55
“Long-Lasting Battery Life Using both ASUS Super Hybrid Engine technology and an energy efficient Intel Atom processor, the Eee PC offers up to 6.5 hours* of battery life…”
“* Tested with MobileMark 2007 in reader mode…” ***NOTE*** IN READER MODE!