If geeks love it, we’re on it

Sempron 2800+

Sempron 2800+

Supplied by AMD


The leading edge of technology doesn’t have to have a bleeding edge price. Not every PC buyer wants to or needs to pay big bucks for performance. AMD introduces Sempron to meet the performance needs of the budget buyer and just when you thought Socket A was dead; it isn’t. Sempron isn’t two processors. Rather it is the adaptation of two technologies based in Socket A and Socket 754. To understand what Sempron is…is to understand who Sempron is for.

sempron_2800_ws02

sempron_3100_ws01

The budget conscious consumer

A vast majority of the PC consumer market is comprised of the lost souls seeking a decent performance PC at a value price. Those two words, value and performance, are going to come up a lot when the word “Sempron” is used. It’s the Sempron drum the AMD marketing folk are banging hard lately. “Thump!”….Value….”Thump”….Performance. The pocketbook rules the budget conscious consumer and while the sexy video cards, gobs of high-end ram and top-of-the-line processors may beckon…the wallet contains moths at the till. The magic number for the thrifty-minded PC buyer, at present, seems to be “around $500.” (USD). That consumer wants value for their money and Sempron is positioned as a value processor.

The dictionary defines the word value as:

An amount, as of goods, services, or money, considered to be a fair and suitable equivalent for something else; a fair price or return.

A sub-$100 processor was almost unheard a few years ago unless it was far down the performance line; outdated technology. Sempron debuts 5 new processors under $100 and the flagship of the Socket 462 Sempron family is just three dollars over. Sempron for Socket 754 comes in at $120.

Processor Price*
Sempron 2200+ $39
Sempron 2300+ $45
Sempron 2400+ $61
Sempron 2500+ $74
Sempron 2600+ $85
Sempron 2800+ $109
Sempron 3100+ (Socket 754) $126

* in USD 1k units

Budget conscious consumers also typically purchase a complete system rather than just a processor so another piece to the puzzle would be the cost of the entire PC. The following is a value system configuration with lowest prices taken from www.pricewatch.com on the date of this article.

Motherboard ASUS A7N8X-VM/400 (LAN/AUDIO/VIDEO) $80
1 x 256 MB Generic PC2700 Memory $23
Maxtor 80 GB hard drive $54
PC case and 300 watt power supply $30
Generic Keyboard and Optical Mouse $20
17″ CRT Monitor KDS 17inch CRT Monitor, Black model XF-7BK $125
Floppy drive $10
Samsung 52x combo drive $36
Altec Lansing 2.1 Speaker system $23
Subtotal without processor, tax and shipping $401

Even With the top end Sempron 2800+ the total purchase price is just a few dollars over $500 USD.

AMD’s apples to apples

AMD changed how it rated its processors a while back. The consumer has always been conditioned to measure a processor’s capability by its GHz rating; more is better. This is what INTEL continually trumpets but even the slightest bit of research will show that more GHz isn’t necessarily better when it comes to comparing INTEL against AMD processors. Admittedly the AMD PR Rating system isn’t as straight-forward when it comes to comparing against basic INTEL GHz numbers. More is better in the AMD product family but what about outside the product family? How, right there on the store shelf, could a consumer be able to compare a 2800+ to a 2.2 GHz INTEL or a 3800+ Socket 939 to an INTEL 3.2 GHz processor?

They can’t without a little reading or asking a few questions.

AMD have gone and attempted to solve this roadblock to a certain extent. At this point is where the conspiracy theorists will find their niche. The fact that AMD have apparently and arbitrarily revised the PR Rating system will be the fuel for their fire. However, the rain cloud putting out the conspirator’s flames is quite simple to see. Sempron for Socket 462 is a Thoroughbred core. AMD is not remarking used stock. They have simply adjusted the Thoroughbred core frequency to better match up against Celeron-D thus the giving them the ability to change the PR rating. AMD have adapted inexpensive, current and proven technology to make Sempron.

Processor
Sempron 2200+ Celeron 2.2 GHz
Sempron 2300+ Celeron 2.3 GHz
Sempron 2400+ Celeron 2.4 GHz
Sempron 2500+ Celeron 2.5 GHz
Sempron 2600+ Celeron 2.6 GHz
Sempron 2800+ Celeron 2.8 GHz
Sempron 3100+ (Socket 754) Celeron 2.8 GHz

The pattern begins to emerge starting with the 2200+ directly comparing to the 2.2 GHz processor. Sempron isn’t new though. It isn’t really a first either.

First there was XP

The Socket 462 AthlonXP processor served AMD extremely well. It is, by no means, “old” technology. It is technology that has fulfilled its life according to the marketing department. “New, better, faster” is the order of the day and while AthlonXP has been a valiant sparring partner with the competition’s top end processors it is no longer “sexy” for big spenders or grabbing headlines.

There are, however, many consumers in the marketplace with a Socket 462 motherboard who may not make the financial expenditure to upgrade to a new socket 754, 939 or 940-pin motherboard and processor. More importantly there are the consumers on a budget who are seeking an entirely new PC. Sempron allows OEM system builders to package a performance PC at a budget price. Sempron comes from the proven design that was the Thoroughbred core. Don’t think there is a secret room at AMD where workers are busily peeling the labels off a stockpile of unsold AthlonXP processors.

The Sempron core and socket 462 design is well developed and well established in the marketplace. It’s inexpensive to produce by AMD and inexpensive to support for OEM system builders and retailers. The cost savings are thusly passed on to the consumer. The Sempron is, in layman’s terms, a re-tweaked AthlonXP to replace Duron. Sempron is meant to go head to head with Celeron. The adjustments are to make Sempron better suit the performance equation when it comes to comparing apples to apples or Sempron to Celeron.

So much for the conspiracy theory.

The next step is to not equate “cheap” with poor performance.

Sempron Specifications

sempron_2800_mws_label

sempron_2800_label

sempron_3100_label

Sempron for Socket 462 is:

  • 130 nanometer technology
  • 333 front side bus speed
  • 256 K of L2 cache and 128 K of L1 Cache
  • 1.6 Volts
  • 84 mm squared die size
  • Approximately 37.5 million transistors
  • supports PC1600, PC2100, PC2700, and PC3200 unbuffered memory
Processor Frequency
Sempron 2200+ 1.50 GHz
Sempron 2300+ 1.58 GHz
Sempron 2400+ 1.67 GHz
Sempron 2500+ 1.75 GHz
Sempron 2600+ 1.83 GHz
Sempron 2800+ 2.0 GHz

Sempron for Socket 754 is presently only available as the 3100+ (1.8 GHz):

  • 130 nanometer SOI technology
  • 64-bit Integrated memory controller up to 3.2 GB/s
  • Single link HyperTransport Link up to 6.4 GB/s I/O for a effective data bandwidth of 9.6 GB/s.
  • 1600 MHz System Bus
  • 256 K of L2 cache and 128 K of L1 Cache
  • 144 mm squared die size
  • Approximately 68.5 million transistors
  • supports PC1600, PC2100, PC2700, and PC3200 unbuffered memory
Processor Frequency
Sempron 3100+ 1.8 GHz

Sempron shouldn’t exactly be thought of as a recycled AthlonXP. It wasn’t too long ago that AthlonXP delivered heralded performance numbers that, in today’s world now known as Sempron, could put respectable performance in the hands of the budget minded consumer.

When Sempron is put side by side with its INTEL counterpart then a dramatic value difference is revealed.

Processor
Price*
Processor
Price**
Sempron 2200+ $39 Celeron 2.2 GHz $72
Sempron 2300+ $45 Celeron 2.3 GHz $78
Sempron 2400+ $61 Celeron 2.4 GHz $68
Sempron 2500+ $74 Celeron 2.5 GHz $88
Sempron 2600+ $85 Celeron 2.6 GHz $88
Sempron 2800+ $109 Celeron 2.8 GHz $122
Sempron 3100+ (Socket 754) $126 Celeron 2.8 GHz $122

*in USD 1k units. **lowest boxed retail www.pricewatch.com date of article.

Benchmarks.

There’ll be a pause to explain the benchmarking procedure before the flames start firing in. Sempron wasn’t due to launch until mid-August. Short-Media only found this out 48 hours prior to the new launch date of July 28th. 2004. Our goal was to get a budget buyer’s perspective of Sempron to publication. It is important to put Sempron into the proper perspective when examining the slew of benchmarks from this and other publications.

To that end there was very little time to benchmark Sempron let alone comparative systems. Two weeks would have been tight and two days was just silly. To that end we pulled data from our archives that was as close a comparison as possible. When looking at the benchmarks remember that the INTEL system is a P4 2.4 GHz processor with Hyperthreading enabled and not a Celeron which is the INTEL comparison processor to the Sempron. The P4 uses PC3200 XMS memory while the Sempron uses PC2700 from Infineon. The P4 is tested with the RADEON 9100 IGP onboard GPU, a FX5200 and GeForce4 MX440 video card. We made a decision to release this type of benchmark information that we had available at press time. It will be our mandate to follow up with a more equivalent comparison and in-depth analysis of Sempron 3100+. We know the benchmarks are not a true comparison and advise you to take that fact into careful consideration when examining the results.

It’s pretty obvious that the odds are stacked against Sempron in this benchmark comparison. It is additionally hanicapped by the onboard MX440 GPU and lesser memory. Given all that it doesn’t quite make for a fair fight. It’s too bad for the P4 as we tried to give it all the help we could.

The test systems.

  • AMD Sempron 2800+
  • ASUS A7N8X-VM/400 Motherboard with integrated GeForce4 MX GPU
  • 2 x 256 MB Infineon PC2700 Memory
  • LG 8x DVD+/-RW.
  • 80 GB Seagate Hard Drive
  • Samsung 950p 19″ Monitors
  • PS/2 Keyboard and Mouse
  • Retail HSF
  • AMK SX1000 modded PC case (window, fans, cables, loom)
  • Enermax 465 Watt FC PSU
  • Windows XP Professional Service Pack 1 updated DX90.b installed.
  • Intel P4 2.4 GHz (HT enabled)
  • MSI RS3M-IL mATX motherboard
  • Integrated ATI Radeon 9100 IGP
  • NVIDIA GeForce4 MX 440-SE Video Card
  • FX5200 Video Card
  • 2 x 256 MB Corsair PC3200 DDR RAM in DIMM 1 and 2
  • Sony 52x CD
  • Western Digital 80 GB Hard Drive
  • Samsung 950p 19″ Monitors
  • USB Keyboard and Optical Mouse
  • Retail HSF packaged with processor
  • AMK SX1000 modded PC case (window, fans, cables, loom)
  • Enermax 465 Watt FC PSU
  • Windows XP Professional Service Pack 1 updated DX90.b installed
  • Intel P4 2.4 GHz (HT enabled)
  • Gigabyte GA-8TRS300M mATX motherboard with RADEON 9100 IGP
  • 2 x 256 MB Corsair PC3200 DDR RAM in DIMM 1 and 3
  • Sony 52x CD
  • Western Digital 80 GB Hard Drive
  • Samsung 950p 19″ Monitors
  • USB Keyboard and Optical Mouse
  • Retail HSF packaged with processor
  • AMK SX1000 modded PC case (window, fans, cables, loom)
  • Enermax 465 Watt FC PSU
  • Windows XP Professional Service Pack 1 updated

Programs used

All tests were run at default video card settings with VSYNC disabled. Anti-Aliasing
and Anisotropic Filtering was left ticked for application preference. AGP aperture
was set to 128 MB. Windows visual effects
was set for ADJUST FOR BEST PERFORMANCE.

Individual performance will vary with any particular or specific timings or
tweaks enabled by you. A 1024 MB page file was moved to D: partition. Temporary
Internet files moved to J: partition at end of drive. OS installed to C: and
programs installed to E:. All programs were benchmarked with initial monitor
settings at 1024×768@75Hz. Your own mileage may very.

Aquamark3

Aquamark3 is a newer benchmark from Massive Development. For the most part
it is a DirectX 8.1 benchmark though it is run with DirectX 90b installed. Four
measurement sets were used. The first has high and low detail with Anti Aliasing
and Anisotropic filtering turned off. The second has high and low detail with
Anti-Aliasing (6x) and Anisotropic filtering (16x) set at max.

aquamark_offofflow

aquamark_offoffhigh

GL Excess

GL Excess is an OPENGL benchmark that is optimized for DX8.1.

glexcess

Quake III high quality

Quake III continues to hang around. This benchmark is one that
most can’t just let go of and it retains grandfather rights in the community.
Many of today’s games are based upon the Quake engine. It wasn’t too long ago that we thought topping 100
FPS was fast. Now we sit at over 300 FPS with the screen set to a high resolution
and detail.

quake

Serious Sam

sam

UT2003 Flyby

ut2003

Wolfenstein Enemy Territory: Railgun timedemo

Wolfenstein Enemy Territory uses an improved version of the heavily
modified Quake III engine from Return to Castle Wolfenstein. The Railgun time
demo results were recorded.

wolfenstein

X2 Rolling Demo

X2 – The Threat is a teaser with a benchmark option for Egosoft’s
upcoming release. It does not use pixel shaders.

x2

Specviewperf 7.1

SpecviewPerf measures the 3D rendering performance of systems
running under OPENGL.

specview

Sisoft Sandra 2004

sandra_math

sandra_multi

sandra_memory

Benchmark Conclusions

Interpretting the benchmarks is made more difficult by the lack of a true and fair comparison. The Intel P4 2.4 GHz processor is aided by the stronger video cards and higher frequency ram. In spite of that the Sempron, “limping” along with the PC2700 RAM and the integrated GeForce4 MX GPU puts up a surprising fight. In some benchmarks the Sempron actually placed first. It’s quite easy to make the leap of faith to extrapolate that Sempron could easily outrun its equivalent Celeron competition. The “unfair” comparison actually revealed the strong performance of Sempron.

3DMark03, Splinter Cell and UT2004 were unable to properly execute on the ASUS A7N8X-VM/400 platform and were not included. This is the fault of the GPU rather than the CPU. Sisoft Sandra reveals the memory advantage of the P4 processor equipped systems.

SpecviewPerf, QuakeIII and Wolfenstein showed highly surprising results with the Sempron’s processor superiority pulling it into top position.

In most of the gaming benchmarks the Sempron placed behind the P4 2.4 GHz HT enabled processor equipped with the superior video cards. What is to be noted is that, despite that handicap, it never finished last.

Conclusions

sempron_2800_ws01

Who is Sempron for? Sempron is a processor that is value priced to be highly attractive to the budget buyer. This isn’t an enthusiast processor but it is no slouch. Sempron is, after all, a Thoroughbred that has been adjusted to better match up against Celeron-D. Enthusiasts have already experienced Thoroughbred AthlonXP, own a superior XP model or have moved to higher ground.

Sempron is value. Sempron is performance.

It’s obvious that AMD are setting their sights on the market of budget-minded PC buyers. What AMD have done is given new life to a veteran processor. For those knowledgeable of the AthlonXP it’s no good to try to compare AthlonXP 2800+, for example, with Sempron 2800+. They are different processors in frequency; 2.25 GHz versus 2.0 GHz. This may not matter because, in a short while, the lower model numbers of AthlonXP will disappear starting with model numbers below AthlonXP 2700+.

What AMD have cleverly done is to give themselves a little bit of “elbow room” for Sempron. The lower frequency and adjusted PR model numbers mean that future Sempron processors can be released for the budget market. This path will also eventually end but not as soon as many had thought. Furthermore, this will give time for Sempron Socket 754 to become fulfill its intended purpose which is the same as Sempron Socket 462. It’s all about the time needed for transition. Socket 462 and 754 will eventually fade away but it’s too firmly entrenched at the moment to be ignored. OEM system builders and retailers appear to believe there’s still a market to support it. The overwhelming amount of available Socket 462 and 754 motherboards supports this.

It was expected that Sempron would be better than Duron but not impressive. As it turns out Sempron put up a strong fight against a P4 (Celeron-D processor were not available from INTEL for review comparison). If Sempron can hold its own against a P4 then it’s easy to assume the outcome against Celeron.

Sempron isn’t a miracle processor though but it could turn out to be one of AMD’s smarter moves. Sempron is a re-tweaked Thoroughbred. The Thoroughbred lived its first life battling it out with top-end P4 processor and now, like a Phoenix from the flames, it rises to battle it out with Celeron at the budget PC level. When consumers learn that they can get “more bang for the buck” with Sempron it can only benefit AMD.

Finally there’s the need for proper perspective. Sempron isn’t intended for a high-end gaming or performance workstation PC. For that theres Socket 939 and 940. Sempron is intended for that “everyday” PC used to send email, play the occasional game, surf the Internet and do homework. Why spend $1000 or more when all that’s needed including monitor can be had for “around $500″? The benchmarks show that Sempron, even on a system with a low-end, on-board GPU peforms as well as or better than a better equipped P4.

sempron_3100_ws01

The Sempron buyer stands to benefit from previous enthusiasts who came before. Sempron is the proverbial Phoenix from what will be AthlonXP’s ashes. When AthlonXP 2800+ hit the market in October of 2002 it launched at almost $400 USD in 1k quantities and it made quite the performance splash. Now Sempron 2200+ launches at $39 and Sempron 2800+ at $109. The performance splash is still there but in time it will be much bigger when you consider the value.

Our thanks to AMD for
their support of this and many other sites.

Highs

  • Inexpensive
  • Great performer
  • Wide and proven motherboard support
  • Lots of support products
  • Proven

Lows

  • Availability

Comments

  1. MediaMan
    MediaMan Little background for you here. Sempron wasn't due to launch until August 18th. There was PLENTY of time to review this product. So I thought. 9AM July 26th I got notification that the launch had been bumped up a tad...to the 28th....OF JULY!!! 0 to 60 on a processor review in 48 hours?!?!

    :eek2:
  2. primesuspect
  3. mmonnin
    mmonnin Thanks AMD for the review.

    Good job Doug!!

    Nice price on some of the low CPUs.
  4. MediaMan
    MediaMan Note a price discrepency on versus our information from AMD on the press release and the AMD website It's a 6 dollar difference across the board. I'll inquire.
  5. GHoosdum
    GHoosdum Doug, I'm really impressed by your quick turnaround time on this.

    And I'm impressed by the Sempron's performance there...
  6. MediaMan
    MediaMan I was surprised too. There are, of course, different reviews out there with different results. Remember that across the reviews hardware configuration does change but one thing is clear...Sempron delivers a better bang for the buck compared to Celeron-D.

    I've yet to get full into the 3100+ and further bench the 2800+ for an even better analysis. There just wasn't time. The benchmark numbers still had the wet paint sign on as the review went on-line.

    We all knew what a good performer the Thorughbred B core was/is. Sempron is that core so while it may no longer wage a good battle at the top end against the top-of-the-line P4s...AMD has just re-assigned it to another front...the battle against Celeron-D.
  7. PressX
    PressX Nice review... looks better than I thought.
  8. MediaMan
    MediaMan Prices adjusted to reflect AMD Processor Pricing. For those of you who saw the other prices; distributor pricing. Not much of a markup...about $6.
  9. Unregistered
    Unregistered One of the worst reviews I've seen for a while. What was the point of using different graphics card (and even integrated ones!) in different setups and then trying to compare the results? This review tells absolutely nothing about the performance level of Sempron... Truly amateurish work.
  10. GHoosdum
    GHoosdum
    Unregistered said:
    One of the worst reviews I've seen for a while. What was the point of using different graphics card (and even integrated ones!) in different setups and then trying to compare the results? This review tells absolutely nothing about the performance level of Sempron... Truly amateurish work.
    Your comment makes no sense. CPUs can be, and often are, benchmarked independently of the GPU all the time. Furthermore, the release of the Sempron was done early, and MM had to struggle to get the article done in two days' time - he admits it himself, and even states that the article will be rewritten once he gets some systems together for straight-up comparisons.

    There's no need to be so spiteful just because the P4 looks inferior... Do you work for Intel or something?
  11. Dexter
    Dexter Great review under pressure, MediaMan. Ignore the unregistered comment as a drive-by insult: the person does not even have the decency to register as a user and tell us who he/she is. Apparently they don't understand the process of a review, which normally requires comparisonsa and benchmarks under different sets of circumstances. :rolleyes:

    Dexter...
  12. TheBaron
    TheBaron
    Unregistered said:
    One of the worst reviews I've seen for a while. What was the point of using different graphics card (and even integrated ones!) in different setups and then trying to compare the results? This review tells absolutely nothing about the performance level of Sempron... Truly amateurish work.
    its a budget cpu, it damn well should be tested against budget graphics chipsets as well - I.E. integrated gfx
  13. Thrax
    Thrax
    Unregistered said:
    One of the worst reviews I've seen for a while. What was the point of using different graphics card (and even integrated ones!) in different setups and then trying to compare the results? This review tells absolutely nothing about the performance level of Sempron... Truly amateurish work.
    Truly amateurish method of stating your opinion. Anonymous trolls.
  14. dragonV8
    dragonV8 I just laugh at people who hide behind a cloak of a anonymity. People like that make me look better. ;D ;D

    Good write up MM. :thumbsup:
  15. Unregistered
    Unregistered The unregistered commentator is back.

    Do you always register when you visit a site for a first time? I don't. It would just take too much time and effort to do in every site I visit (especially when there is no easy way to do it instantly). So it's not hiding, just pure common sence.

    And no, I don't work for intel. I actually am bit of a AMD fan.

    Anyway, why does this review suck? Well, it doesn't answer the question that this kinda review should always answer: How fast is the Sempron? There are just too many variables. Out of the 9 test, 8 measure the overall performance of the system, and only one pure proseccor performance.

    Ofcourse it's nice to find out the real life overall performance of system, but when compared systems differ from each other so dramatically (=different graphics cards), there just is no way to tell how much a proseccor effects the result, and what are the effects of the other parts used. What makes this review especially bad is that 8 out of 9 test had this mistake.

    This test might aswell been nForce2 IGP vs. Radeon 9100 IGP test (or actually it would have been more usefull that way). If you don't have time to do a proper review (which would tell something about Semprons speed) it's better to leave the review undone alltogether. Releasing this kinda review just makes me (and undoubtably some others aswell) avoid the site in the future...
  16. MediaMan
    MediaMan Let me address a few points that the guest brought up.

    1) We allowed for unregistered comments to encourage people to comment. Some users hesitate at signing up for anything. I wanted to give site visitors more opportunity to comment be it good or bad.

    2) The "dragstrip of benchmarks"

    What is "real world performance?" It is a subjective interpretation. I always encourage any visitor to read our reviews and at least 2-3 other reviews from our colleague's sites. The big picture will give a far more accurate interpretation.

    Sempron is for a budget buyer. Those buyers typically do not purchase top-end video cards. They buy a "Dell" on a price point. Those PCs have low end graphic cards or integrated graphics cards. Budget buyers don't go out and spend $500 on a video card. They spend $500 on the whole system.

    Sempron is for that marketplace.

    I agree with the unregistered guest that the comparison wasn't perfect. The review had a full disclaimer at the first benchmark page to explain the tight turnaround and the reasons.

    There'll be a pause to explain the benchmarking procedure before the flames start firing in. Sempron wasn't due to launch until mid-August. Short-Media only found this out 48 hours prior to the new launch date of July 28th. 2004. Our goal was to get a budget buyer's perspective of Sempron to publication. It is important to put Sempron into the proper perspective when examining the slew of benchmarks from this and other publications.

    To that end there was very little time to benchmark Sempron let alone comparative systems. Two weeks would have been tight and two days was just silly. To that end we pulled data from our archives that was as close a comparison as possible.


    If the unregistered guest does not want to visit this site in the future due to this then I am sad to lose a visitor. Comments...even negative comments that are posted without malice or poor language are constructive feedback.

    The unregistered guest did bring up a very important point.

    How fast is the Sempron? There are just too many variables.

    The overall message isn't about how fast Sempron is. It's, as stated quite clearly, the value/performance equation. If you want "fast" then purchase the top end AMD or INTEL products. This is the Achilles Heal of reviews. It all comes down to the pretty graphs where a visitor can quickly, without wasting a neuron, see how "fast" one product is compared to another. Add up the pretty charts and then make a decision to buy or not.

    That isn't the best way to buy but it is what readers expect.

    The second part of the unregistered guest's comment was There are just too many variables.

    This is too true. All of us don't buy the same components. We mix and match. We overclock and we don't. We use different OS and software configurations.

    That is why I always endorse and push people to read our reviews and two to three others to get a better overall picture. I get the same hardware in as other review sites. Show me two or three out of the dozen review sites who got the same numbers...

    If the hardware is the same...shouldn't the numbers be?

    Again...I welcome the unregistered guest's comments; present and future...good an bad. If this particular guest thinks that this review "sucks" then so be it. If they do not visit again then I hate to lose a visitor and it reminds me to work all that much harder to keep future visitors coming back.
  17. Madball
    Madball And how does it overclock? Are they locked?
  18. Dexter
    Dexter
    Unregistered said:

    Out of the 9 test, 8 measure the overall performance of the system, and only one pure proseccor performance.
    Around here, we like to call them processors.

    If you are going to be so hypercritical of a review of a processor, it least learn how to spell it, dude :thumbsup:

    Dexter...
  19. Thrax
    Thrax You have to allow everyone a typo, Dexter. It's obvious from the rest of his post that he's rather literate.
  20. Dexter
    Dexter
    Thrax said:
    You have to allow everyone a typo, Dexter. It's obvious from the rest of his post that he's rather literate.

    I "allowed" a typo: "commen sence." ;D

    But "proseccor" typed twice in his comment...? Nah. Especially when the word was typed correctly 38 times in the article. If the dude read the article so closely that he was able to offer such incisive commentary on it, one would think he would have found the correct spelling of the item he seems to think he knows a lot about. ;)

    Dexter...
  21. Unregistered
    Unregistered First of all it's kinda low to pick on my typos, especially considering that english is only my third language... Secondly, if you want to have a debate about something one of the basic rules is that stick to the subject (and don't get personal).

    But now I have to confess that before I posted my first comment, I didn't read the disclaimer. Instead I just browsed the article trough and got amazed about the conducted tests. And I would suspect that most people read reviews like I did; just browsing them quickly trough and checking the bechmarks (especially if you have red few Sepron reviwes earlier). So my initial reaction might not have been as justified as I tought it would be. Sorry for that...

    But still I still stand behind my two claims
    1. It's pointless to run gaming benchmarks in a processor review , when the limitating factor is the used graphics card. You might aswell used XP3200+ instead of Sempron, and the results would have been virtually same. Ofcourse you can still run couple of gaming test just to see how a typical budget system performs, but 8 out of 9 is still too much.

    2. If the article isn't good enough to be published (and you know it), don't publish it. Like I said, most just browse trough the article disregarding any disclaimers. If you want people to know that you have had the opportunity to review the product, then make a article focusing on something else than somewhat useless benchmarks.

    Hmmm... Perhaps I could register now :)
  22. mmonnin
    mmonnin MM just doesnt publish graphs for people like you to look at, he publishes something to read for the rest of us.
  23. Straight_Man
    Straight_Man
    Unregistered said:
    One of the worst reviews I've seen for a while. What was the point of using different graphics card (and even integrated ones!) in different setups and then trying to compare the results? This review tells absolutely nothing about the performance level of Sempron... Truly amateurish work.
    Something to think about-- value boxes are likely NOT to have top line graphics. As a CPU in isolation article, or a CPU versus CPU-- pure CPU-- article, limiting graphics is maybe not the best approach. BUT, the Sempron is not likely to come on a hyper-high end system otherwise. So, it was real life as Sempron is expected to be bundled and is Targetted by AMD to be bundled-- it is intended to replace Duron and is priced accordingly. How many duron boxes come with the latest separate video card in system???

    For a 48 hour turn, VERY GOOD. As MM said, he will be digging into Sempron more-- expect some good digging with a practical and not theoretical perspective. This was NOT how to stretch to ub3rmost (that is largely ART and takes time to do right), this was a very good first look in a very short notice deadline timeframe.

    Unregistered one, please write a better review if you want.... Let us look at it, please. MM has lots of things he wants to write about, and does NOT do just this site. He's a career media person also. And practical. So he starts with practical scenarios. This is an enthusiast's site, not a pure engineering tech site. So, some folks come as techs, others (majority) come wanting to see what they can expect from a Sempron-centered SYSTEM.

    Over the coming months I would expect to see both from here (first looks plus detailed followups as MM stated would happen)-- not demanding, EXPECTING because of what has come before. EXPECTING GOOD, in other words. But CAREFUL, covering fine detail well, means you take time. Two days to test and write, you end up with a first look. Two-three weeks, you get and can take time to dig and try tricks and apply system art in configuring which the average bear in the woods does not know. Never should have been published??? US mags and sites for enthusiats do first look style things all the time.
  24. Rogan
    Rogan
    1. It's pointless to run gaming benchmarks in a processor review , when the limitating factor is the used graphics card. You might aswell used XP3200+ instead of Sempron, and the results would have been virtually same. Ofcourse you can still run couple of gaming test just to see how a typical budget system performs, but 8 out of 9 is still too much.
    Utter Rubbish, all i wanted to know is how well sempron would run against an equivelent XP chip AND Intel. There is no reason why these should not be shown in benchmarks.

    Unfortunately the only part of your argument that was valid, that all graphics tests should be run using the same graphics card was lost because you trolled. Well played ... dumbass.

    Media Man, you gotta make sure that the GPU matches, I mean you can't always match up motherboards and memory, well you could between XP and Sempron in this case. But swapping graphics cards in a graphics benchmark nullifies your testing. I really wanted to see how the chips stacked up against each other, but as unregistered says (poorly) we cant see because you have swapped the hardware.

    Anyway, thx for the review.
  25. Unregistered
    Unregistered < Not the same Unregistered guest >
    First, I'd like to complement MediaMan on taking the time to write the article in the first place. As a professional analyst, I know the time it takes the gather the material and to produce a readable piece for public consumption. Two days is indeed impressive.

    My main comment though, is that he should have considered using a Celeron based system (which are widely available) instead of a P4. As you have implied in your message, the Sepron is meant as a value alternative and the attraction to the platform will be its price/performance over equivalent Intel-based value systems. While it is impressive that the Sepron fares well against a higher priced P4 system, it would be more valuable to see, dollar for dollar, the incremental improvement of this system.

    And I agree with the other Unregistered guest, hammering someone over spelling errors are rather low.
  26. Thrax
    Thrax Actually, I'm of the opinion that if the Sempron holds its own against the Pentium 4, then it would do even better against the Celeron. That's the real test of value; budget versus enthusiast.
  27. MediaMan
    MediaMan
    Unregistered said:
    < Not the same Unregistered guest >
    My main comment though, is that he should have considered using a Celeron based system (which are widely available) instead of a P4. As you have implied in your message, the Sepron is meant as a value alternative and the attraction to the platform will be its price/performance over equivalent Intel-based value systems. While it is impressive that the Sepron fares well against a higher priced P4 system, it would be more valuable to see, dollar for dollar, the incremental improvement of this system.

    You are most correct. The Sempron, since it is a (sort of) reclassified XP, should have been pitted against the Celeron. The reason I didn't was two-fold. 48 hour deadline and money.

    I was faced with purchasing a Celeron and motherboard to support it. (Some $500 CAD) Contrary to popular belief we don't make wads of cash so the bill would have been footed by me personally.

    If I had more than 48 hours then I would have done this. I could have sold off the Celery and partially recouped my investment. Making a site grow and giving the readers better articles does take that sort of committment. I have a 2" stack of receipts that show quite well that it isn't always "free stuff" :)

    I thank you for your commments. Everything helps me produce better articles for you. Glad that you did take the time to comment...It is appreciated. :)



    /me wanders off looking for his lottery tickets.
  28. Geeky1
    Geeky1
    I have a 2" stack of receipts that show quite well that it isn't always "free stuff"
    Tell me about it... that XP-120 review ended up costing me like $350 (1 DFI LanParty, only to discover it didn't fit the XP-120, 1 ASUS P4P800E-Deluxe, 1 XP-120) :rolleyes:

    But then, I got $350 worth of new toys to play with out of the deal too... ;D ;D
  29. Gobbles
    Gobbles
    MediaMan said:
    You are most correct. The Sempron, since it is a (sort of) reclassified XP, should have been pitted against the Celeron. The reason I didn't was two-fold. 48 hour deadline and money.

    I was faced with purchasing a Celeron and motherboard to support it. (Some $500 CAD) Contrary to popular belief we don't make wads of cash so the bill would have been footed by me personally.

    If I had more than 48 hours then I would have done this. I could have sold off the Celery and partially recouped my investment. Making a site grow and giving the readers better articles does take that sort of committment. I have a 2" stack of receipts that show quite well that it isn't always "free stuff" :)

    I thank you for your commments. Everything helps me produce better articles for you. Glad that you did take the time to comment...It is appreciated. :)



    /me wanders off looking for his lottery tickets.



    I disagree. Using a P4 was just in my opinion as the celeron has never been able to compete with the duron/sempron...
  30. MediaMan
    MediaMan
    Gobbles said:
    I disagree. Using a P4 was just in my opinion as the celeron has never been able to compete with the duron/sempron...

    Which is now how AMD has positioned Sempron so in the argument of which would you choose; Sempron or Celeron...which one would win and be the obvious choice?

    :)
  31. no time to register
    no time to register why bicker instead of research? who cares which CPU plurbs the X axis fastest with the same equipment--within .0006MHz per nansecond per a pretty graph? this sort of review is pratical, a bit biased by the writer and the components, and very effective. excluding neophytes, one can read the review and form an opinion and have a starting point to dig. at this stage can't we all go to our chosen sites and get thespecs, again excluding neophytic box builders, the writer can offer an educated and hopefully reliable opinion on the value vs. horsepower of the CPU because the CPUs have different targets, determining to the Nth degree which is faster kind of skirts the usefulness of the article, doesn't it?
  32. still too busy
    still too busy sorry for the typos befur innywun desides to plug me
  33. primesuspect
    primesuspect you're questioning the usefulness of an article that is almost a year old :wtf:

    This processor, and any arguments surrounding it, are long obsolete.
  34. Nomad
    Nomad I want to see the written writ for exumming this thread's corpse :|
  35. profdlp
    profdlp
    Nomad said:
    I want to see the written writ for exumming this thread's corpse :|
    Ghoul. :zombie:
  36. Gobbles
    Gobbles
    MediaMan said:
    Which is now how AMD has positioned Sempron so in the argument of which would you choose; Sempron or Celeron...which one would win and be the obvious choice?

    :)
    sempron of course. I was saying that pitting a sempron against a p4 is a fair fight as a regular xp would have crushed a p4 and along that lines, the sempron would have crushed the celeron so bad all that would be left under the hsf would have been a small amount of silica dust.

    Im a tad back logged on my replies to threads... ;D :D :thumbsup:

Howdy, ! Got something to say?