Sajan 8-K Other Events

Exhibit 99.1

MathStar, Inc.

Conference Call Script

Quarter ended December 31, 2005

Conference call date: February 28, 2006

Conference call time: 3:30 PM CST

Call Duration 60 minutes

Speaker

Comments

Doug

Good afternoon. I want to start by thanking everyone for joining us this afternoon on our first quarterly conference call. Joining me today are Dan Sweeney, MathStar’s Chief Operating Officer and Jim Cruckshank, our Chief Financial Officer. Since many of you are new to MathStar we will begin with a quick review of our technology and product concept. MathStar has developed a new chip architecture - called Field Programmable Object Arrays or FPOAs — that we believe have substantial competitive advantages over currently available programmable chips. MathStar’s goal is to establish the FPOA architecture as the premier solution in the programmable logic segment of the semiconductor industry FPOAs combine the low development cost, rapid time to market and extended time in market advantages of existing programmable logic devices while delivering the low per unit cost and high performance advantages of ASICs. FPOAs are ideally suited to implement high performance algorithmic functions across a wide spectrum of digital applications — FPOAs have very broad market applicability. Our current FPOA competes primarily with the high performance products marketed by Xilinx and Altera. We believe that compared to FPGAs, our FPOA has substantial long-term competitive advantages of higher performance, faster time to market and lower manufacturing cost. As we expand our family of FPOAs, we believe we can use our inherent competitive advantages to win designs across a broad segment of the PLD market and become the leader in a new category of programmable logic. Because the FPOA is a chip with broad applicability in multiple applications, we believe the FPOA will become an important element in the rapidly growing programmable logic market. Analysts expect this market to grow from $3 billion now to $6 billion over the next four years. During 2005, we made substantial progress in our transition from a technology development company to a customer and market driven

company. I would like to highlight the important milestones of 2005.

• We taped out and delivered to customers a working version of our 1 GHz FPOA which has enabled us to make multiple design wins prior to the release of our production silicon.The 1 GHz FPOA is at least twice as fast as the highest performance offerings from our competitors, its functional level programming makes it easier to program, and its much smaller size enables high gross margins.

• We invested in our corporate infrastructure and attracted a outstanding management team with the experience and ability to execute a rapid growth strategy and scale the business. We established a performance driven organization and disciplined financial controls.

• Retired United States Air Force General Merrill McPeak and Morris Goodwin joined Benno Sand and Doug Pihl on the MathStar board.

• We greatly strengthened our engineering team as we transitioned from development silicon to commercialized production silicon.

• We systematically developed our marketing strategy to focus on key initial growth markets including machine vision and professional video.

• We strengthened our strategic relationships with Honeywell, Valley Technology and Summit Design.

• We brought our total number of early customer design wins to 12 at the end of 2005.

• We expanded our Application Development team and completed eight basic algorithms that can be used as building blocks in customer applications.

• We brought our Company public based on our business strategy of introducing an exciting new technology into one of the most rapidly growing segments of the semiconductor market, with a business model capable of producing high gross margins, and with the right team to execute our plan.

During the remainder of the call we will discuss these items in greater detail, starting with Dan Sweeney with an update on our product development efforts.

Dan.

Dan

Thanks, Doug. The entire MathStar engineering team has been focused on getting production-level FPOA silicon, programming tools and application libraries into our customer’s hands to support the initial revenue ramp. I’ll take some time to review progress in each of the three areas; I’ll start with the status of the FPOA silicon. Since May of 2005, MathStar has been sampling the current revision of FPOA silicon for customer evaluation, software development and early prototyping. Due to design marginalities, this silicon cannot be manufactured economically in volume, and a revision of this silicon is currently in development. We have made great progress and expect to achieve our next major milestone in late April called tapeout. Tapeout occurs when the database that contains the design information meets all of the complex process-specific design rules, has been fully verified and is sent to Taiwan Semiconductor, or TSMC, our wafer foundry, to generate masks. These masks are used by TSMC to manufacture FPOA silicon wafers. Our plans had called for tapeout of MathStar’s first production FPOA in late December 2005; however, as the new engineering team implemented the rigorous design verification processes required to ensure we were taping out an FPOA design we could manufacture in volume, we encountered issues that we hadn’t anticipated. To address these issues, we implemented improvements in the chip design, which included:

• Complete and detailed timing analysis, which drove redesigns to deliver high 1GHz yields;

• Design changes were implemented to ensure short test times; and

• Expanded I/O capability and on-chip memory were added based on customer feedback.

Implementing these improvements will ensure we tape out an FPOA design we can produce economically in volume at the targeted yields. We are now on track to tapeout at the end of April, which will support getting chips in customer hands in late June and early July. Over the course of the 3^rd quarter of 2006, we plan to increase our test capacity and therefore our ability to produce FPOAs in greater volumes.

Software is key to MathStar’s long-term success. We are investing in two critical areas, the software tools our customers use to program the FPOA and FPOA application libraries used as building blocks by

MathStar customers to lower their development costs and speed time-to-market.

In the Tools area, we made significant progress towards maturing the tools our customers use to design and program our FPOAs.

• In December, we extended our relationship with Summit Design to provide Visual Elite™ SystemC-based design libraries for MathStar’s FPOA technology.Included in the new agreement are provisions that will allow MathStar to provide its customers with 90-day free evaluation licenses of the Summit tools together with MathStar’s tools and its new FPOA evaluation board. We believe this will make it easier for customers to evaluate FPOAs as solutions to their system performance problems.

• We have significantly increased the resources supporting tools development, supplementing existing software engineers with contractors experienced in programmable logic tools development.

In April, we expect to begin selling an early version of the new tools that support this upcoming FPOA. This will allow our customers to begin developing their applications for the new FPOA prior to chip availability. In the 2nd half of 2005, MathStar hired a team of application engineers to develop FPOA libraries for customers in our target markets. We expect these libraries of common blocks used by our customers will reduce their development costs and speed time-to-market. We’ve completed the initial set of libraries required for high performance video applications and have begun integration and testing. We expect to complete our first Machine Vision libraries in Q2 of 2006. We are well on the way to having production quality FPOA silicon, our first generation of FPOA programming tools and the initial set of FPOA software libraries for video and imaging applications. I’m excited to be part of the MathStar team that delivers the next generation of programmable logic to customers hungry for a high-performance, programmable solution. I’ll now turn it over to Jim for a financial update.

Jim

Thanks, Dan. For the fourth quarter of 2005, our revenues were $80,000 compared to $30,000 for the same quarter last year. Research and development costs for the quarter ended December 31, 2005 were $2.4 million compared to

$1.7 million for the same quarter last year. The increase of $700,000 was primarily the result of increased payroll and contract engineering costs. Selling, general and administrative costs were $2.2 million compared to $900,000 for the same quarter last year. The increase of $1.3 million was primarily the result of:

• increased payroll costs of about $800,000, of which approximately $500,000 was the non-cash expense associated with restricted stock awards and employee options, and

• increased costs associated with being a public company of approximately $500,000. This includes legal, outside audit costs, insurance, printing and transfer agent costs.

Our cash used in operations and investing activities for the quarter ended December 31, 2005 was $5.6 million compared to $2.8 million for the same quarter last year. The $2.8 million increase was primarily the result of increased R&D expenditures, building the SG&A infrastructure and costs associated with being a public company.

Comparing the third and fourth quarter of 2005, research and development costs increased from $1.8 million in Q3 to $2.4 million in Q4. The $600,000 increase was the result of engineering staff additions and contract engineering costs. These increases are due to a combination of planned additions and additional resources needed to add the capabilities that Dan identified earlier.

SG&A costs increased from $1.5 million in Q3 to $2.2 million in Q4. The $700,000 increase was primarily due to our beginning to recognize the non-cash expense of restricted stock grants of about $500,000 and costs associated with the convertible debt and the Registration Statement on Form S-1 we filed on December 23^rd of about $200,000. We anticipate we will recognize approximately $2.0 million of non-cash compensation expense in 2006 related to restricted stock.

The Company’s cash used in operations and investing activities for the quarter ended December 31, 2005 was $5.6 million compared to $2.3 million for the quarter ended September 30, 2005. The $3.3 million increase in cash used in operations and investing activities was primarily the result of increased R&D expenditures, building the selling, general and administrative infrastructure and costs associated with being a public company.

On a year-to-year basis, 2005 revenues were $134,000 compared to $130,000 for 2004. Research and development costs for the year ended December 31, 2005 were $8.8 million compared to $5.2 million for 2004.

The $3.6 million increase was the result of increased payroll, contract engineering and material costs. Selling, general and administrative costs were $6.2 in 2005 million compared to $3.7 million for 2004. The $2.5 million increase was the result of:

• increased payroll costs of about $1.4 million, of which $600,000 was non-cash expense related to stock options and restricted stock awards, with the remainder being costs associated with building our infrastructure; and

• increased costs associated with being a public company of about $1.0 million. This includes legal, outside audit costs, travel, insurance, printing and transfer agent costs.

Our cash used in operations and investing activities for the year ended December 31, 2005 was $15.1 million compared to $9.3 million for 2004. The $5.8 million increase in cash used in operations and investing activities was primarily the result of increased R&D spending, building the SG&A infrastructure and costs associated with being a public company.

Our cash balance at December 31, 2005 was $20.1 million. Based on our cash balance, cash burn and our expectations about the timing and extent of chip revenue during 2006, we believe we will have enough cash to carry us through 2006. However, we are exploring options to raise additional financing once we have demonstrated full production-ready silicon and have achieved additional design wins. We will be exploring several alternatives, including private or public investments and a secondary offering, and we may consider strategic partnerships. Any of these alternatives may include additional equity or convertible debt offerings. We will disclose additional information when available.

We made great progress during 2005 and early 2006 in administration and finance. We completed our public offering in October of 2005, we anticipate being ISO 2001 certified later this year and Sarb-Ox compliant ahead of the regulatory requirements.

I will now turn the call back over to Doug.

Thank you, Jim.

We are also committed to the accomplishment of a number of significant milestones in 2006.

The first priority of our entire staff is the successful completion of the full production version of the FPOA. The entire team has committed to

assuring a fully production-ready part, we have taken the time to get it right, and we are highly confident we have achieved that.

Our Marketing organization, under the direction of Sean Riley, is actively engaged in building awareness of our new technology and presenting the FPOA to the industry through presentations at conferences and in technical publications. By the end of 2006, we expect that many industry analysts will recognize the FPOA as a new programmable logic alternative.

We finished the quarter ended December 31, 2005 with twelve design wins. The Company defines design wins as customers who have purchased an FPOA development board and tools. While we don’t expect all design wins will turn into production revenue, we are focused on developing long-term relationships with our customers. Initial wins are with customers that require the high performance of the FPOA, making them more willing to adopt a new architecture. We believe that once these customers achieve a new performance level and competitive advantage, they will have a greater incentive to award us additional design wins. Our team is focused on our customer’s success with these initial wins.

Our new VP of Worldwide Sales, Glen Wiley, started in early January. He brings with him many years experience in the semiconductor industry, most notably building sales organizations to drive rapid growth. We are well on our way to establishing an organization consisting of direct Area Sales Managers, or ASM; Field Application Engineers, or FAE; and independent manufacturers’ representatives. The manufacturers’ rep model will enable us to maximize the exposure of the FPOA in our targeted commercial markets while minimizing cash burn. Our FAEs and ASMs will help close the opportunities the manufacturers’ reps identify. We are pleased to announce that, as of February 28^th, four manufacturers’ reps have committed to carry the MathStar FPOA in their product offering. We are getting very positive traction in our target markets — machine vision, professional video and test and measurement. In order to accelerate the design win ramp, we plan on adding manufacturers’ reps in all major North American markets by the middle of this year.

We believe we are on track to close a minimum of 25 design wins during 2006. Our marketing and sales strategies have produced a strong pipeline of potential customers at various levels of commitment. We will provide updates on our progress at the end of each quarter.Due to a number of factors, not the least of which is the availability of production silicon, we expect a significant number of those design wins to occur in the latter

half of the year, and we are encouraged by a number of new factors affecting our ability to win designs:

• Production silicon samples should be available in late June and early July. This, in particular, will help us drive design wins and the revenue ramp.

• In the 2nd quarter, we expect to begin to make the tools software containing the models for our new chip broadly available so customers can start their development efforts prior to silicon being available.

• In the 3^rd quarter, we plan to have a new evaluation board with a PCI system level interface available to customers. This will allow customers to use a standard PC to evaluate the FPOA.

• As Dan mentioned, we will also be able to provide 90-day free Tools evaluation licenses to customers, eliminating the need for customers to invest in tools in order to evaluate our FPOA.

• The combination of these last two factors should allow us to get evaluation boards in customers’ hands at a much lower price. This will enable a larger number of customers to evaluate our FPOA and reduce the time from initial engagement to design win.

• We will have completed some very impressive application developments that we will be able to demonstrate to customers. These are applications that will clearly show the dramatic benefits of the FPOA technology.

MathStar has the clear opportunity to be the segment-defining player in programmable logic. We look forward to a very successful year and believe a year from today we will have demonstrated significant market penetration and progress in making FPOA the next generation of programmable logic. We are focused on building long term, profitable growth for our shareholders.

That concludes our formal presentation. We have about 15 minutes remaining for questions.