Talk:Pentium 4

Brand-name vs. code name for chip vs. code name for microarchitecture

1. "Pentium 4" is a brand name, so are "Core 2 Duo", "Core i3-xxxx", etc.
2. Nehalem (microarchitecture) is a code name for some microarchitecture
3. Clarksfield (microprocessor) is a code name for some chip/die/package fitting into some CPU socket

This distinction does not only make sense, it is also most relevant, because e.g. there are chips marketed under the "Pentium 4"-brand that support the x86 (32-bit) instruction set and others which implement x86-64 (64-bit) instruction set. Additionally there are a bunch of extensions like – e.g. MMX, SSE4 – that are implemented by some and not by other processors marketed under the same brand. One could market very different things under the same brand: Blue Digshit seems to still be available. User:ScotXW^t@lk 13:59, 7 December 2017 (UTC)

Appropriate?

Overclockers mistakenly believed that the Northwood was a better choice for overclocking, while in reality the Prescott would outperform the Northwood and generate less heat in overclocked conditions. ^that's still generally found to be false, atleast for the early revisions of prescott. it wasn't until prescott improved a fair bit as far as manufacturing went where that was true as initial prescott chips didn't clock as high as Northwood —Preceding unsigned comment added by 76.87.89.227 (talk) 04:04, 8 March 2008 (UTC)

Ram types and the Pentium 4

Does anyone know what types of RAM work with the P4? What's the relationship between FSB speed and RAM? Should this article link to others describing this relationship?

Mr. Jones 11:33, 24 Sep 2004 (UTC)

THe P4 has supported PC100 SDRAM, PC133 SDRAM, PC800 RDRAM, PC1066 RDRAM, DDR266 SDRAM, DDR400 SDRAM, DDR2-266 SDRAM, DDR2-400, and perhaps a few I've missed. Last Avenue 00:14, 25 January 2006 (UTC)

Usually the FSB clock is the same as the RAM clock. For example, Intel's current 200MHz FSB corresponds with the RAM clock, 200MHz. DDR400 runs at 200MHz but is called DDR400 because of DDR.

Some types to add: PC-600/PC-700 Rambus, all DDR-1 RAM types (PC-1600, PC-2100, PC-2700 and PC-3200), at least all available DDR-2 Ram types released until now. Memory support always depends on the motherboard used, they usually support only one type of RAM. --Denniss 03:09, 25 January 2006 (UTC)

Up to Nehalem (microarchitecture) the memory controller was part of the North bridge. I also marked it here: Template:Intel_processor_roadmap. So look up the appropriate chip set and that article should tell the supported RAM-type. The socket-article should also tell. Or look it up in the list: List of Intel chipsets User:ScotXW^t@lk 12:25, 7 December 2017 (UTC)

Dates Right?

Processor Cores>Northwood> "A 2.4 GHz P4 was released in April 2002, and the bus speed increased from 400 MT/s to 533 MT/s for a 2.53 GHz part in May, 2.6 and 2.8 GHz parts in August, and a 3.06 GHz Pentium 4 arrived in November."

These dates dont seem right to me. If anyone has sources or can back these up. Im pretty sure you could get 2.4Ghz P4's well before April 2002.

No comment about the date, however this statement in the main document needs to read: "A 2.4 GHz P4 was released in April 2002, and the bus speed increased from 400 MT/s to 533 MT/s for a 2.26 GHz, 2.4 GHz, and 2.53 GHz part in May, 2.6 and 2.8 GHz parts in August, and a 3.06 GHz Pentium 4 arrived in November." --I personally own a 2.26 533 FSB and I know a 2.4 also existed in a 533 FSB.

I'm right now looking at a Pentium(R)4 labeled Intel (R)(C) '01. 2.40GHz/512/533 SL6sH. Intel website (https://ark.intel.com/content/www/us/en/ark/products/27438/intel-pentium-4-processor-2-40-ghz-512k-cache-533-mhz-fsb.html) "launch date Q1'02". "May" isn't Q1, so the May date given here may be wrong. It doesn't seem to be referenced.

It still seems like a long lead time, but I guess Intel didn't 'release' the 2001 parts until April/May. — Preceding unsigned comment added by 121.200.27.15 (talk) 08:13, 25 August 2023 (UTC)

Or maybe the "release" date is a retail release date? — Preceding unsigned comment added by 121.200.27.15 (talk) 08:23, 25 August 2023 (UTC)

Removed text

I removed, at least for now, the following PR blurb:

In addition to making significant increases in performance from the Pentium III class of CPUs. The Pentium 4 produces low amount of heats compared to its AMD counterparts, and features a silver electroplated sheet of copper ontop of the core to protect it. A retention mechanism which placed the physical load of the cooling mechanism on the motherboard via a retention mechansim mounted on holes through the motherboard, this is important as shearing off of the retention tabs on the CPU socket. is a common problem when some heatsinks. In addition to producing less heat, each version version of the core for Pentium 4 undergoes testing to determine its maximum safe operable tempature, heat output, maximum safe core voltage. The Pentium 4 also includes safeguards to make use of the information gathered about the maximum safe operable speed of the Pentium 4. If the Pentium 4's on core thermal diode reads higher than the maximum safe operable tempature, it will descrease the core speed to reduce output. This is very important, as even if a Pentium 4s cooling mechanism somehow becomes detached, it will survive the incident.

Quite a bit needs to be done with this to correct some misconceptions.

• One needs to be careful about making "significant increases in performance" generalisations. The thing about the P4 is that it ticks very fast, but does relatively little work per tick. In terms of IPC, it was a major step backwards. In terms of raw clockspeed it was (eventually) a major step forward. The original P4s were, frankly, slugs. They got better - much better - when Northwood arrived.
• The "low power, low heat" thing is nonsense. Compare, for example, the P4 1.5GHz and the (slightly faster) Thunderbird 1000c - 52W at full load vs 54W - a very marginal difference indeed. (I'll try to find time to dig out a few other examples later tonight.) Essentially, provided you compare similar process technology parts that have similar computational power, there is very little difference between P4 and Athlon. AMD claim that the Athlon is more power efficient than the P4, though I think that they maybe take it a little too far.) (The P-III, on the other hand, is remarkably power efficient. A 1.8 micron P-III 1000 - which was the equal of a P4 1.5 and not far behind the Athlon 1000c, performance wise - consumed just 33W.)

Damn! I'm out of time - I'll come back and finish this off a little later tonight. Tannin 09:49 Jan 30, 2003 (UTC)

OK, more power consumption figures, as promised. First, I have to apologise for my mistake earlier tonight. I said that power consumption is, in fact, roughly the same between the Athlon and the P4. I was wrong. I've been checking my facts, and the Athlon clearly uses considerably less power than the P4 to do the same amount of work. The numbers above compare the maximum consumption of the Thunderbird 1000 to the typical consumption of the P4. The correct figures are:

54W: 0.18u Thunderbird 1000C

75W: 0.18u Pentium 4 1.5 (Socket 423)

(There were at least three variants of the 1.5 - the other two are 76W and 79W.)

Now let's consider the next performance level. There was hot debate as to which of these two chips was, in fact, faster, the P4 1.7 or the Thunderbird 1400. Most experts - about two out of three - regarded the Athlon as the faster of the two, but some argued the other way, and everone agrees that they were very close.

72W: 0.18u Thunderbird 1400

87W: 0.18u Pentium 4 1.7 (Socket 423)

Now let's look at the next step again (and another hotly debated "which is the best of two rather similar chips" contest. The consensus is that the XP 1700 was comfortably faster than a P4 2.0 - remembering that this is the old 256k 1.8 micron P4, not the vastly improved 0.13u Northwood that came along in January 2002.

64W: 0.18u Athlon XP 1700+

96W: 0.18u Pentium 4 2.0 (Socket 423)

100W: 0.18u Pentium 4 2.0 (Socket 478)

Just for fun, let's throw in the Northwood as well - but we must remember that in doing this we are not comparing like with like, as the 2.0 Northwood is faster than an XP 1700 and made on a different and much cooler running process. I'll also include the 0.13 Athlon at 2000+ for comparison

69W: 0.13u Pentium 4 (Northwood)

62.8W: 0.13u XP 2000 (Thoroughbred)

Finally, let's take the two current-model 2600 parts:

68.3W 0.13u Athlon XP 2600+

81W: 0.13u Pentium 4 2.6 (Northwood)

Next: A retention mechanism ... this is important as shearing off of the retention tabs on the CPU socket is a common problem when some heatsinks. Sorry, but that's pure PR bull. In my 20+ years of working with PCs, I have seen this problem in the flesh exactly once. It was toward the end of last year, about October, I think, and the CPU in question was, of all things, a Cyrix 686. We devised an ugly but effective workaround of the sticky-tape and string variety, but then the customer decided it ws time he had an upgrade anyway, so we just threw the motherboard away. At the time it was about four or five years old and worth maybe US$10. Never seen it happen to an Athlon or a P4.

It has happened to me. The retention tabs on the Socket 370 snapped off, this happened twice. It was on a Via Apollo 133T (694T), MSI Model 6318 Version 5. I was using the stock Retail Pentium III 1.13ghz fan.Patcat88 17:19, 25 Jan 2004 (UTC)

Each version version of the core for Pentium 4 undergoes testing to determine its maximum safe operable tempature, heat output, maximum safe core voltage. Yup. Quite right. Each version of every other CPU manufactured in the last ten or twenty years also. Not to mention vido card GPUs, RAM, and anything else that uses large transistor counts.

If the Pentium 4's on core thermal diode reads higher than the maximum safe operable tempature, it will descrease the core speed to reduce output. This is very important, as even if a Pentium 4s cooling mechanism somehow becomes detached, it will survive the incident. And here, at last, we finally strike gold. This is an important feature of the P4 design, and (unless someone gets in first) I'll return this to the entry. It's very different to the Athlon. The P4 clock throttles (i.e., slows itself down) in a user transparent way. The only way you become aware of the fact that your P4 is overheating is when you notice that it's not performing properly. (By the way, this aspect of the P4 design makes benchmarking very difficult - unless you go to a lot of trouble with cooling, you never get the same result twice running.) In contrast, the Athlon delivers 100% of its rated performance 100% of the time. When the Athlon's thermal diode hits a pre-set critical temperature, it cuts the power. It's a diference of design philosophy: the Intel part is prepared to sacrifice performance to avoid an unintended shutdown, the AMD is prepared to risk an unintended shutdown in order to deliver 100% performance. Neither chip can self-destruct unless the technician is grossly incompetent. (Note that older model Athlons - in particular the Thunderbirds - relied on motherboard manufacturers to provide effective thermal sensing. Often, they didn't. I always regarded that as a design error on AMD's part - personally, I wouldn't trust the average motherboard manufacturer to tie his own shoelaces 10 times out of 10. Tannin 15:40 Jan 30, 2003 (UTC)

Ack! Please don't throw perfectly good Socket 7 motherboards away. Lots of people would be happy to have them Crusadeonilliteracy 14:18, 16 Jan 2004 (UTC)

Intel Cores

I made a list of intel cores, there are some not included in this article though. I dont know much about them but maybe someone can include some information? Read the Inqurier article at the end, it is very good.

List of Intel cores

This is a list of Intel microchip 'cores'. A core can be considered a chip design.

• Willamette
• Netburst core
• Prescott
• Tejas
• Dothan
• Northwood

Intel faces performance struggle for two hard years, The Inquirer

--ShaunMacPherson 01:21, 16 Oct 2004 (UTC)

Netburst is not really a core, it is the underlying architecture for all P4 variants. Dothan is a Pentium M core that should be treated in the respective article. --Echoray 12:42, 16 Oct 2004 (UTC)

Disambig?

I noticed that Willamette redirects here. Do you think we should add a disambig notice on the top in case people want the Willamette River or Willamette University?--Clipdude 08:15, 9 Nov 2004 (UTC)

NPOV

Also of concern was the fact that reviews showed it took a 5.2 GHz Prescott core to match the performance of an Athlon FX-55 [1]. Considering Intel boasted at launch the Pentium 4 architecture was designed for 10 GHz operation, this can be seen as one of the most significant, certainly most public, engineering shortfalls in Intel’s history.

This is not NPOV or correct - a 5.2 Ghz Prescott will soundly beat an FX-55 in multimedia encoding, as can be seen in the Mp3 encoding graph where the 5.2ghz Prescott beat the FX-55 by 35%, and the 4ghz Prescott beat it by 6%. The same applies for video encoding. The review only shows the Prescott has to be clocked at 5.2 Ghz to beat the Prescott in games and WinRAR compression. The next two paragraphs are also highly opinionated. --Defsac 02:01, 18 Apr 2005 (UTC)

Willimette

"Many observers concluded that the Thunderbird was still faster overall, but the performance gap was sufficiently narrow that it was not unreasonable for partisans of either camp to claim superiority. For Intel, this was a very significant achievement. The firm had held the x86 CPU performance crown for nearly 16 years straight, with only two brief exceptions prior to the release of the AMD Athlon."

I disagree with this statement, as it only refers to the SD-ram utilisation of this CPU, where'is if used with RD-ram, was far superior to the rival offerings of the time. Infact, it took a whole 2 yrs untill AMD and Intel itself to better the raw long-code performance of an RDRAM chipset/cpu of a bygone era, even then only just matched it, and it wasnt untill a futher revision by both companies that it was bettered. The above reporting shows how deaply the bias was against RAMBUS, and the lenghts some reporters went to to discredit them.

Uhh, the Willamette was never a superior CPU. It didn't have enough cache, or clock speed, to compete with Athlon, especially when Athlon XP arrived. Pentium 3 could run with most Willamette systems. Of course, SDRAM versions of Willamette were even worse off because they were simply crippled by the lack of bandwidth (Netburst loves bandwidth). [2] --Swaaye 06:09, 1 June 2006 (UTC)

Mobile Pentium 4

There in no information at all on the Mobile Pentium 4 processors (both Mobile Pentium 4 and Pentium 4 M). Some info can be obtained from Mobile Pentium 4 and Pentium 4 M product information pages. Those processors were widely used before Pentium M was launched. -- Eagle^amn 13:39, September 4, 2005 (UTC)

I just added some info on the P4 mobile chips, hopefully someone can add to it and help make it clearer. There aren't a lot of good resources on these chips, still trying to find some release dates, phase out dates, processor speeds etc... this page helped me some. Any help would be welcome! --Fxer 16:46, 7 October 2005 (UTC)

dual core

under the dual core section, it says presler is based on the 65nm version of smithfield. To be more clear it should say the 65nm single cores of smithfield are called cedar mill,

Dual Core

Some vandal made a comment about how the L2 cache bottleneck is not eliminated by providing each core with an independent cache, which is completely untrue. He also tried to assert that the Presler will not be a "true Dual-Core" chip, but rather a Multi Core chip. Well, unless Intel magicly ships presler with three or four cores (not damm likely), the distinction between dual-core and multi-core is pointless; I corrected the paragraph to reflect reality more closely.

Unsourced opinions

I removed this comment:

In retrospect, the Pentium III core was technologically superior to Pentium 4, which can be seen as an example of what happens when marketing teams draw up technical project specifications.

This is a common opinion (Pentium 3 was superior) combined with rather derisive speculation (marketing did it).

The first needs a source and the second needs to go—unless a more informed source can also be found for that one. 82.92.119.11 23:29, 18 January 2006 (UTC)

Why the Prescott ended up in such a disaster can be attributed to internal politics at Intel. The marketing department wanted ever higher clock speeds, to differentiate their products from AMD. Processor design was dictated by marketing needs, rather than architectural requirements. Careers were then built on the concept of higher clock speeds, and the termination of the P4 project when it finally came, had consequences for many members of the management team at the well-funded desktop division.

Is there a source for this? I'm in the unfortunate situation of not being able to get a reliable source for this in order to include it in a report that is to be peer-reviewed. If anyone knows, please contact me via my talk page. I think it's probably fair to say that these comments fairly reflect what people think happened, so I'm sure it could be modified to be acceptable even if a great source can't be found. Important questions: Who made the decisions? Who got fired? The story makes sense, but without a reliable source and some firmer details this text can't be left as is. Gantlord 11:11, 8 December 2006 (UTC)

In his book "The Pentium Chronicles" Bob Colwell, one of the Willamette chief architects wrote that he left Intel after he realized it was a mistake to bet on ever higher clockspeed but couldn't make a dent into the product roadmaps. He says something like "Willamette was not fun". Unfortunately, I don't have the book at hand anymore. --Echoray 12:26, 8 December 2006 (UTC)

Thanks echoray, I also turned up this if people think it's interesting: http://www.bit-tech.net/bits/2006/08/17/marketing_blunders_of_our_time/1.html Gantlord 13:04, 8 December 2006 (UTC)

Rendering Errors

I have rendering errors in internet explorer on the main page. The icon of the P4 obscures some of the text. Could this be changed? (i will try myself).--BorisFromStockdale 20:28, 19 January 2006 (UTC)

Best Processor

come on, unpack those loops, first make function compounding, bit processing style then aply the info to b processed, use that pipeline n that predication for something get into 6th generation microchips able of evn more than 1000 instr/1T n most important:dont mention me

Florin Matei — Preceding unsigned comment added by 93.118.212.93 (talk) 08:29, 10 March 2012 (UTC)

What is the best proccesor out there?

Depends. Intel Core 2 is the best for multitasking and its great for laptops because it consumes little power, same goes with the original Intel Core. The Pentium 4 is great for video encoding and single-core applications because of its raw power, brute force and high clock speed but its not very efficient and runs VERY hot (not suitable for laptops) although some of the problems were fixed with Cedar Mill. The P4 is okay at multitasking because it has hyperthreading. The Pentium D is like two P4's stuck together so its better than the P4 multitasking but otherwise very similar (still hot and inefficient). The Celeron and Pentium dual core are weak but very cheap. So it depends on what you want to get out of your system. The Intel Core2 runs at a slower speed but its multicore so when your running two programs, the second core will kick in and youll be able to do your work two times faster. Its also based on PIII so its more efficient. It also has alot of cache and a faster front bus speed. However, the P4 will beat it in single-threaded applications because of its higher clock speed (the most important part of a processor). You also need to take into account the other components, such as RAM size/speed and Hard drive speed. 203.184.7.28 (talk) 22:55, 20 December 2007 (UTC)

Technically the Core 2 is based on the Pentium M, which is based on the PIII. Your comment about the clock speed being the most important part of a processor is wrong. The Core 2 still beats the P4 even with its lower clock speed because it can do more work in each clock cycle. The P4 was designed to have a high clockspeed (since "MHz sells"). This lead to there being a lot of heat produced, making the Core 2 necessary. Overall, you want a Core 2 (or something based on Core), NOT a P4 (or P4 based chip). (In AnandTech's Core 2 review, there were several benchmarks where the cheapest C2D at launch (E6300) outperformed the previous P4 based flagship CPU (PD 965)) Also, although RAM size does have a significant effect on performance as well as the RAM size (although that is a smaller effect), the hard drive speed does not have any noticable effect on performance. -- Imperator3733 (talk) 17:04, 21 December 2007 (UTC)

List of Intel Pentium 4 microprocessors

I've recently created a List of Intel Pentium 4 microprocessors page in the same vein as the lists of AMD processors. I think there are enough P4s that a list page is justified, and I've never been comfortable with the idea that AMD processors get list pages and Intel processors don't. Right now, it's a subpage of my user page as I'm considering it a work in progress. After filling in some more stuff, it should be ready to be moved into the main namespace. I'd also like to hear any thoughts/comments/concerns before I do so as well. Jgp 11:05, 1 April 2006 (UTC)

Looks good. Maybe replace the FSB MHz with MT/s (Megatransfer). This shoudl also be done for the AMD listings using the wrong MHz. --Denniss

Removing the table

Since List of Intel Pentium 4 microprocessors exists, I'd like to propose deleting the huge table on this page, due to the redundancy of it and the fact that the list page is better-organised. Since it's a pretty big section, I'd like to get some consensus before I remove the table. Does anyone else think removing the table is a good idea? jgp 04:29, 30 May 2006 (UTC)

The table here looks provides an easier look at the processors without having to look at all of the individual speed bumps. It is quite redundant, but I think the smaller table here will give the reader a better look at the Pentium 4 upgrades, rather than looking through the other page. —Last Avenue [talk | contributions] 05:13, 1 June 2006 (UTC)

leave the table alone. —Preceding unsigned comment added by 88.218.27.9 (talk • contribs)

why? jgp ^T_C 19:41, 30 August 2006 (UTC)

because the info is useful. if you add them in this article http://en.wikipedia.org/wiki/List_of_Intel_Pentium_4_microprocessors then you can remove it —Preceding unsigned comment added by 88.218.27.9 (talk • contribs)

That "info" is a cluttered mess. It's inaccurate and unmaintainable. This is a prose article and only prose descriptions belong here. jgp ^T_C 13:35, 31 August 2006 (UTC)

Without going over it with a fine-tooth comb, nothing sticks out as inaccurate. The "only prose" justification, taken literally, is extremely weak; articles like Sun benefit greatly from the summarization of information in a tabular format, and loads of articles use some form of tabular infobox (including processor sockets) to great effect. I think what you really mean is that it's too much table for a WP article. And perhaps it really isn't appropriate (though it's a great table, frankly). But I would suggest you hash this out here instead of just deleting it over and over. — Aluvus ^t/c 14:32, 31 August 2006 (UTC)

Added: The Pentium 4 introduced new technological advances such as hyper-threading, a feature to make one physical CPU appear as two logical and virtual CPUs. Evidence: Look up hyper-threading on www.techweb.com/encyclopedia

Overclocking

Any idea, who made 4ghz or on laptop, i've heard they released 4ghz but never to the public? also 70% faster comapred to 12% intel, something on main page, what's that

Production dates

I have noticed that there is mention of the Pentium 4 line being retired on July 27, 2006. In the introduction. It was also added to the infobox - this information is not cited. AFAIK Pentium 4 CPUs are still being produced and production of Pentium 4s will not stop until next year. Can anybody confirm the so called 'retirement' of these chips? - MaXim 05:53, 1 November 2006 (UTC)

It seems to me that's the day they retired it, but i added the the date they were retired but it got reverted for some reason--Falcon866 21:59, 6 January 2007 (UTC)

Maybe you got reverted because the P4 is still in production ?!? --Denniss 01:19, 7 January 2007 (UTC)

Pentium 4 line being retired on July 27, 2006 then how are they still in production--Falcon866 13:25, 7 January 2007 (UTC)?

Not retired yet, surpassed by Core 2 but production lines still live. --Denniss 21:32, 7 January 2007 (UTC)

Case closed, the Pentium 4 line is discontinued. I am a computer technition so I would know. Our store is no longer able to get Pentium 4 chips at all. Some of the people here may be confusing the Pentium 4 line with the dual core Pentium D. The Pentium D is still widely avalible as a low end, cost effective dual core solution. The single core Pentium 4 however is not being produced any more. Production on the Pentium 4 chips was halted on July 27, 2006 when the Core 2 line appeared. Jdlowery 03:41, 12 March 2007 (UTC)

Please provide a source for the P4 end of production. It was for sure not ended once the Core2 appeared. There are still lots of single core Pentium 4 (Cedar Mill) available, at least in Europe. --Denniss 18:58, 12 March 2007 (UTC)

Fastest one core processor

There are mistakes on this page, the processor 3.8 should beat 2.8 by amd or at least equal, are you sure you have things correct? Can somebody check on this, something's rotten in denmark... And somebody in europe claims they are selling 4ghz laptops? Damn, they should have just made it 4ghz and call it a day.

The 4GHz Pentium 4 580 could be preorderd by at least 4 resellers in Germany (for a premium of 1400-1500€, well above the 1000€ Extreme Editions), but none claimed to have them in stock. It was also never official announced by Intel. There were however samples distributed to hardwaretest-sites. I remember one test on tomshardware.de (.com probably as well) that introduced the 580 and the 3,73EE. The 580 finished with noticeable more performance, so it should qualify as fastest single-core modell ever produced (though not marketed) in multithreaded or SSE heavy benchmarks. (Which would have placed an "ordinary" Pentium 4 above the permium-ExtremeEdition in power and price)

Unfortunately, those tests were removed with all other traces of the 580, so these accounts are not referenceable to the standards of wikipedia. However google still finds one News anouncing the listing at various dealers:

www.computerbase.de/news/hardware/prozessoren/intel/2005/juli/intel-pentium-4-580-mit-4-ghz-gelistet

It also includes a modellnumber (BX80547PG4000E), which is mentioned on various other websites (but got deleted from Intels lists as well, as all other traces of the CPU).

ruyven_macaran 13:15, 25 October 2010 (UTC) —Preceding unsigned comment added by 188.108.182.107 (talk)

Prescott section

The Prescott section has grammar errors and is hard to read. (Anon).

I have heard that the Prescott design reference from Intel specifies a grilled side panel and ducting. Many companies offer such products which are marketed as 'Precott ready', 'Precott compliant', etc. I cannot find the papers about this but if anyone can find them and post them here I'll write it in and try to address the grammar issues raised by Anon above.--ChrisJMoor 21:04, 1 April 2007 (UTC)

I agree, the Prescott section is horrible. Someone has spammed it with blogs (e.g. niche hardware/OC sites) and opinion. To maintain neutrality, it should cite mainstream, respected hardware review sites. And try sticking to the point a bit more. Denniss seems bent on reverting improvements. (Anon). —Preceding unsigned comment added by 134.134.137.71 (talk) 15:20, 5 March 2010 (UTC)

Fair use rationale for Image:Intel Pentium 4 HT.png

Image:Intel Pentium 4 HT.png is being used on this article. I notice the image page specifies that the image is being used under fair use but there is no explanation or rationale as to why its use in this Wikipedia article constitutes fair use. In addition to the boilerplate fair use template, you must also write out on the image description page a specific explanation or rationale for why using this image in each article is consistent with fair use.

Please go to the image description page and edit it to include a fair use rationale. Using one of the templates at Wikipedia:Fair use rationale guideline is an easy way to insure that your image is in compliance with Wikipedia policy, but remember that you must complete the template. Do not simply insert a blank template on an image page.

If there is other other fair use media, consider checking that you have specified the fair use rationale on the other images used on this page. Note that any fair use images uploaded after 4 May, 2006, and lacking such an explanation will be deleted one week after they have been uploaded, as described on criteria for speedy deletion. If you have any questions please ask them at the Media copyright questions page. Thank you.BetacommandBot 08:02, 5 June 2007 (UTC)

I'm handling it. Suigi 15:44, 5 June 2007 (UTC)

Northwood was faster than Athlon

http://www.computerbase.de/artikel/hardware/prozessoren/2003/test_intel_pentium_4_240c_ghz_athlon_xp_2800/5/ —Preceding unsigned comment added by 134.155.99.42 (talk) 06:24, 27 November 2007 (UTC)

thermal junction

"Pentium 4-M had a thermal junction temperature of 100 degrees, in other words, about 40 degrees more than the desktop Pentium 4" - I don't think this correspondences to the reality. IMHO it should be correct "case temperature" instead of "junction". The junction temperature for semiconductors uses to be much higher - for classical silicium transistors up to 200°C. -- 8ohmSpeaker (talk) 19:17, 29 December 2010 (UTC)

Virtualization at Cedar Mill

"It [Cedar Mill] has a 65 nm core and features ... Virtualization Technology." Is that right? 77.11.112.98 (talk) 05:05, 18 August 2011 (UTC)

the 6th generation processing chips, the way i c it :)

did i say 1000 here: o course i did: only their beeep(the c......n word, sorry) might went wrong then — Preceding unsigned comment added by 93.118.212.93 (talk) 21:41, 25 May 2012 (UTC)

come on, unpack those loops, first make function compounding, bit processing style then aply the info to b processed, use that pipeline n that predication for something get into 6th generation microchips able of evn more than 1000 instr/1T n most important:dont mention me

Florin Matei — Preceding unsigned comment added by 93.118.212.93 (talk) 08:29, 10 March 2012 (UTC) — Preceding unsigned comment added by 93.118.212.93 (talk)

About a possible N->log(N) time optimization

basically, microprocessor instruction r function of 1 bit of 2 binary variables... if that possibly means that any such function, no matter how complex in the background, is formally depending only of 16 variables, then, computer might gain experience that can be useful in n->log(n) time optimizations such as FOR loops especially factorials or PI the most far decimal place known.We assume that we make a difference between the function n the information 2 be processed.

Thank You for Your Time, Florin, Romania — Preceding unsigned comment added by 93.118.212.93 (talk) 07:47, 6 September 2012 (UTC)

Sudden Northwood Death Syndrome (SNDS)

The article mentions that early stepping Northwood cores can suffer from Sudden Northwood Death Syndrome (SNDS).

Which Northwood cores are affected by SNDS? The earlier ones without HT technology (those that run at FSB400 and FSB533)? Or are FSB800 Northwood cores (those with HT technology) also susceptible to SNDS? — Preceding unsigned comment added by 222.165.42.62 (talk) 17:13, 28 June 2012 (UTC)

Class action

There is a class action for certain Intel Pentium 4 processors. Does anyone think this should be included in the article?

http://www.intelpentium4litigation.com/

                                             Printphi (talk) 23:20, 27 October 2014 (UTC)

yes! absolutely! it's the reason someone would want to read about Pentium 4 on wikipedia. Thomas144 (talk) 20:18, 4 November 2014 (UTC)

"Intel eventually added XD Bit (eXecute Disable) and Intel 64 functionality to Prescott"?

Intel 64 or x86-64 is an 64-bit architectural extension, rather than some additional instruction sets, it could not later easily be put onto some a microarchitecture when the number of transistors did not even happen to change obviously. The very first Intel Pentium 4 Prescott processor released in February, 2004, and the very first Pentium 4 Prescott with EM64T released in August, 2004, the same year! Or in other words, Pentium 4 Prescott is just an x86-64 processor, but only enabled with such related functions on select models, rather than later or eventually added such things, they had already presented there, but just disabled. — Preceding unsigned comment added by 221.9.13.186 (talk • contribs) 11:00, 22 May 2017 (UTC)

The above seems to be just discussion of the writer's unreferenced claims and assumptions. It is not clear what point you're trying to make, but talk pages are for discussing improvements to the article, and any changes to the article will require references to reliable sources - not your opinions, assumptions, and conclusions therefrom. Jeh (talk) 01:52, 24 May 2017 (UTC)

brand

Pentium 4 is a brand. CPUs branded Pentium 4 were available for a couple of Sockets, and even more important, some Pentium 4-branded CPUs were 32-bit only, while others were 64-bit. User:ScotXW^t@lk 12:26, 7 December 2017 (UTC)