5 Research Questions for the Term Paper.
CIS-512
Andrew Dick Dr. Hamid Gooda

1. The first question of the five is about Advanced Micro Devices (AMD) processors. AMD is known for their processors, however, their line of products that they have stems even further now. In the microprocessors division they have the Athlon variety, X2, K10, X2, AMD Opteron (Miller, 2010), are the specific microprocessors that they have. They also offer servers. The processors of the Opteron are the processors that are available. AMD also offers motherboards to their users as well The AMD Crossfire series are available in this line along with many others. AMD recently acquired a leading video card manufacturer and is promising more advanced video capabilities coded into new versions of their motherboards. It also offers software and memory solutions for businesses. AMD is said to offer the same basic product as Intel, however, at a cheaper price. Intel is established as the market leader in producing motherboards and processing chips for personal computers, and their prices and development have influenced AMD from the start (Miller, 2010), AMD actually produces "clones" of Intel products as well as their own specific line of chips and motherboards as explained above. The AMD Athlon XP processor runs very close to an Intel 4 processor and it is about half the price (Miller, 2010). For example, if a side-by-side comparison is used between the AMD Athlon 64 X2(939) and the Intel Core 2 Extreme 775. The Intel processor specifications and price are: QX6700 / 2.66GHz Socket 775 1066MHz $1059.99 QX6700 / 2.66GHz Socket 775 1066MHz $1059.99. The AMD processor and specifications are: 4200+ / 2.20GHz Socket AM2 1000MHz (2000 MT/s) $82.99 X2 BE-2300 / 1.90GHz Socket AM2 1000MHz (2000 MT/s) $89.99 X2 4400+ / 2.30GHz Socket AM2 1000MHz (2000 MT/s) $92.99 (Miller, 2010). As it is easy to see the AMD is very close in comparison, more significantly, the price is far less for the AMD. With that being said, AMD does offer Intel compatible processors. These processors are the exact same pin design so that it can be used with the same motherboard without needing a custom motherboard design. As one can see, the performance is compatible between AMD and Intel, leaving AMD as a good choice for hardware computer needs for a better price than Intel. The second question is pertaining to IBM’s holographic storage systems. The nature of holographic storage is the ability of quick retrieval and storage of huge amounts of data. This is all possible due to lasers that can store "pages" of electronic patterns within a volume of special optical materials instead of just the surface like with magnetic hard drives. In traditional holography, each viewing angle gives a different view of the same object. In holographic storage, instead of presenting another view of the same object, a different page of information is presented. This type of retrieval speed makes this extremely vital to primary and secondary storage. Up to 10,000 pages (each with one megabit of information) have been stored in a crystal the size of a sugar cube. And because there are no mechanical moving parts and all the information in each "page" is accessed simultaneously. All stored data can be input or read (accessed) very rapidly; a speed of one billion bits a second has already been demonstrated.
2. Currently, holographic storage is used for high quality videos, and images for application in a variety of industries. Such industries are telecommunications, aerospace, medical, The primary reason for all of these industries using this is due to the ability to perform essentially immediate data searches through huge digital libraries by simply illuminating the media with all of the stored information (a holograph) with a pattern of the requested information (Schechter, 2010)
3. The majority of current ASR algorithms rely to some degree on pattern matching, speech consistency (or similarity) is also important. The literature reviewed explores the relationship between ASR recognition performance as measured by accuracy (% of words correct, divided by total number of words used) and the severity of dysarthria as measured by intelligibility, various sources of speech variability, and perceptual measures of intelligibility and consistency. There are 2 methods used in classification and summarization of ASR. They are; In the first method, the FE is enriched with a pitch detection algorithm. The pitch feature (or a related feature describing the harmonic structure of the spectrum) is used to create a separate ‘acoustic models’ (using e.g. a pitch/delta-pitch codebook). The acoustic model used in tests is a combination of the gross-spectral model and the pitch model. In this way, one can improve the performance of a recognition system for Chinese with about 10-30 percent reduction in word error rate. The ‘tones’ are to be transcribed in the lexicon on the syllable level. The second method is to evaluate the pitch in the FE, but to use the pitch information only at the stage of rescoring the N-best list that is produced by the regular ASR recognizer. In this way, one may use pitch, word stress or other supra-segmental or lexical information to improve the recognition score. In contrast with method 1, this method can also be used when pitch is to mark information that is on the supra-word level, e.g. focus. Therefore this or a similar method is particularly useful in improvement of recognition results in dialogues (Blaney, Wilson2000).
Early work on the Switchboard conversational telephone corpus by Blaney et al suggested that topic ID from 50-60% accurate ASR output can do as well as topic ID from human transcripts. Their test set consisted of 120 conversations, evenly split among 10 topic labels. The topic models (as differentiated from the ASR acoustic models) were trained on the human transcripts, rather than recognition output.
One of the biggest growing uses of such technology is trademarked by Ford Motors. They have a system inside the dashboard called Sync. This takes the spoken language or words and converts them into actions that the car can use. For example, If an individual wanted to listen to a specific cd which they had in the player they could say “Elvis Presely” and if they had an Elvis Presley album in the player the car would in return put that album on.
4. Both of these methods are very inexpensive now. At one point the price of the CD’s and DVD’s was very high making the ability to archive media wither it was pictures, music, or any other data very pricey. The fastest performance is DVD-ROM drives. Even though they have a much lower RPM (revolutions per minute) value, data transfer rates are substantially higher than a CD-ROM drive at equivalent RPMs, because the data is compressed by the use of a greater number of smaller data pits and a smaller track pitch (the distance between tracks). For example, a 1X DVD-ROM drive transfers data at 1,250KBps, whereas a 1X CD-ROM drive transfers data at only 150KBps. By 1998, multispeed DVD-ROM drives became available that were capable of reading DVD media at double-speed, resulting in a transfer rate of 2,700 KBps, and of rotating CDs at 24X. By the end of that year DVD drive speeds had increased to 5X. Presently, DVD drives are capable of a 6X speed (8,100 KBps) for DVD media and 32X speed for reading CD-ROMs (Schechter, 2010). Both offer an estimated life span 75 years. However the ability to have a media reader to be able to read such media in 75 years is highly unlikely. For example, most current machines cannot even read a 5.25 floppy discs, which were very popular about 10 years ago. So the concept that these discs will last that long is not ideal. I honestly would not choose either. I have not been a fan of either of these 2 technologies mainly due to the fact that external hard drives no matter the size have become inexpensive. This leads me to the final question for the number 4 question. External hard drives are a much better way to store not only media, but anything needed that can be moved or kept in a safe location, leaving in my opinion; both DVD and CDs obsolete (Edwards, 2006).
5. A NAS with those specifications would cost roughly about 5,000. A SAN under those same specifications would be around 8,000. A SAN is in most cases double the price of a NAS. A SAN has far more functionality and performances than a NAS has, thus making it a better choice for higher storage access performance. NAS is much easier to configure and administer. Due to its main function of just being a large or mass storage device that multiple pieces could and will have access to. This makes setting up a NAS far easier and much easier to maintain do to the simplicity of its function and design.

Work Cited
Blaney, B., & Wilson, J. (2000). Acoustic variability in dysarthria and computer speech recognition. Clinical Linguistics & Phonetics, 14(4), 307- 327.
Burd, S. D. (2011). Systems architecture (6th ed.). Cengage.
Edwards, J. (2006) "Distinguishing an Archival Grade Optical Medium." Imation, Inc. Presentation to the 2 February 2006 meeting of the Government Information Preservation Working Group (GIPWoG).
Miller, M. (2010, June 09). Amd, intel begin 2011 chip combat. Retrieved from http://www.pcmag.com/article2/0,2817,2364855,00.asp

Schechter, B. (2010). Leading the way in storage . Retrieved from http://domino.watson.ibm.com/comm/wwwr_thinkresearch.nsf/pages/storage297.html